The Thoughtful Integration of Mediation Into Bilateral Treaty

The Thoughtful Integration of Mediation into Bilateral Investment Treaty Arbitrationi

Nancy A. Welsh2 and Andrea Kupfer Schneidera

ABSTRACT While the current system of investment treaty arbitration has definitely improved upon the "gunboat diplomacy" used at times to address disputes between states and foreign investors, there are signs that reform is needed: states and investors increasingly express concerns regarding the costs associated with the arbitration process, some states refuse to comply with arbitral awards, other states hesitate to sign new bilateral investment treaties, and citizens have begun to engage in popular unrest at the prospect of investment treaty arbitration. As a result, both investors and states are advocating for the use of mediation to supplement investor-state arbitration. This Article draws upon

1. This Article extends ideas begun in Nancy A. Welsh, Mandatory Mediation and Its Variations, in U.N. CONFERENCE ON TRADE AND DEV., INVESTOR-STATE DISPUTES: PREVENTION AND ALTERNATIVES TO ARBITRATION II, U.N. Doc. UNCTAD/ WEB/DIAE/IA/2010/8 (Susan Franck and Anna Joubin-Bret, eds., 2011), available at http://www.unctad.orglen/docs/webdiaeia20108-en.pdf [hereinafter INVESTOR-STATE DISPUTES II], Andrea Kupfer Schneider, Using Dispute System Design to Add More Process Choices to Investment Treaty Disputes, in INVESTOR-STATE DISPUTES II and Nancy A. Welsh & Andrea K. Schneider, Becoming "Investor-StateMediation," 1 PENN ST. J. L. & INT'L AFF. 86 (2012). We thank Jack Coe, Susan Franck, Mariana Hernandez Crespo, Anna Joubin-Bret, and Jeswald Salacuse for their careful comments and advice on this Article. We also express thanks for the insights and comments offered by colleagues at Penn State University, Dickinson School of Law, and participants at the colloquium hosted by the Penn State Journal of Law and InternationalAffairs, the University of Minnesota-American Society of International Law Conference on International Economic Law, and the Joint Symposium on International Investment Law and Alternative Dispute Resolution at Washington and Lee University School of Law. Finally, we extend special thanks to Katie Rimpfel, as well as Katie Lonze, Mark McGill, Katherine Parr, Kevin Schock, and Nida Shakir, for their excellent research assistance. Any mistakes are, of course, our own. 2. William Trickett Faculty Scholar and Professor of Law, Penn State Univer- sity, Dickinson School of Law. 3. Professor of Law and Director, Dispute Resolution Program, Marquette Uni- versity Law School. Harvard Negotiation Law Review [Vol. 18:71

dispute system design principles, the socio-psychological research and theories regarding procedural justice, and the U.S. experience with court-connected mediation. Using these lens, the article examines the models of mediation that have been shown to be effective, the importance of ensuring that mediation offers something different from the other procedural options available to resolve investor-state disputes, and the mechanisms that increase the likelihood that disputing parties and stakeholders will perceive individual outcomes and the larger system as fair. The Article also examines the U.S. domestic experience to identify the elements of the mediation process that can be, and have been, made compulsory and the effects of this choice, as well as different approaches for ensuring the quality of the mediation process and its accountability to the disputing parties and other stakeholders. Ultimately, the Article recommends the integration of a default model of mediation into the investor-state context that begins in a facilitative manner, in order to increase the likelihood of trust-building and information exchange regarding important underlying interests, but also permits both evaluative interventions by the mediator and discussion of relevant legal norms. The Article further concludes that if stakeholders' input is sought and considered regarding mechanisms for the referral of disputes to mediation, some elements of mediation could be made compulsory. More specifically, the dispute resolution clauses in investment treaties could require the parties to participate in an initial meeting to discuss the potential use of mediation or other consensual procedures, with the parties themselves then choosing whether to proceed further, when, and with what process. Finally, the Article recommends the establishment of a small pool of well-known and well-respected investment treaty mediators who will offer a reasonably strong and pragmatic guarantee of quality in the short- term and engender a heightened perception of trust in the process. These mediators should also possess the temperament and skills to provide the default model of mediation. In the long term, however, evaluation and mentoring must be put into place to permit thoughtful cultivation of both the model of mediation that is best suited for the investor-state context and the next generation of mediators.

CONTENTS

Introduction ...... 74 II. The Current State of Investor-State Arbitration and Concerns About Mediation ...... 78 Spring 2013] Bilateral Investment Treaty Arbitration

A. The Emergence of Problems in Investor-State Arbitration ...... 78 B. Adding Mediation to the Mix ...... 82 C. Critiques of Mediation-and the Challenges to be M et ...... 84 III. Dispute System Design and Procedural Justice ...... 91 A. Dispute System Design ...... 91 B. Procedural Justice ...... 95 IV. The Mediation Process, Referral Schemes and Quality Control Mechanisms ...... 105 A. The Mediation Process ...... 105 1. Basic Differences Between Mediation and Arbitration ...... 105 2. A More Nuanced Appreciation of the Relationships Among Models of Mediation, Arbitration, and Other Processes ...... 108 3. Recommended "Default" Model of Mediation for the Investor-State Context ...... 116 4. The Relationship Between the "Default" Model and States' Preventative Initiatives .... 117 B. Mediation Referral Schemes ...... 120 1. Scope of Compulsory Mediation ...... 123 2. Degree of Intrusiveness of Compulsory Elem ents ...... 124 3. Parties' Input Within Compulsory Systems .... 126 4. Mixed Compulsory and Voluntary Elements... 127 5. The Effects of Compulsory Mediation ...... 128 6. Recommended Compulsory Elements in the Investor-State Context ...... 132 C. Mediation Quality Assurance/Accountability M echanism s ...... 133 1. Accountability of the Parties to the Court and Each Other ...... 134 2. Accountability of the Mediators to the Court and Parties ...... 135 3. Accountability of the Courts to the Parties and the Public ...... 137 4. Recommendation for the Investor-State Context ...... 139 V . Conclusion ...... 142 Harvard Negotiation Law Review [Vol. 18:71

I. INTRODUCTION

Investment treaty arbitration has evolved over time to encourage and protect foreign investment. Historically, companies that invested money in foreign countries hoped that their governments would step in and protect them if the host government did not treat them well. In some instances, this "gunboat diplomacy" not only protected investment but also resulted in violent confrontation. As part of the effort to move away from such power-based relations among states and toward cooperation within international organizations, member governments of the World Bank ratified the Convention on the Settlement of Investment Disputes Between States and Nation- als of Other States in 1965 and established the International Center for the Settlement of Investment Disputes ("ICSID").4 While the multiple attempts since then to create a multilateral treaty to govern international investment have resulted in failure, the bilateral investment treaty ("BIT") model has thrived.5 Patterned after friendship, commerce, and navigation treaties (FCNs), BITs were originally designed to provide for the protection of foreign investment in each country and for treatment that is no different from that experienced by host country investors.6 BITs also now provide for arbitration under the rules of ICSID or the United Nations Commission for International Trade Law ("UNCITRAL"). 7 Arbitration proponents urge that arbitration has played a significant role in easing global commerce by: (1) enabling states and investors

4. See INT'L CTR. FOR SETrLEMENT OF INV. DISPUTES, ICSID CONVENTION, REGU- LATIONS AND RULES 5, available at https://icsid.worldbank.org/ICSID/StaticFiles/ basicdoc/CRREnglish-final.pdf; Andrea K. Schneider, Getting Along: The Evolution of DisputeResolution Regimes in International Trade Organizations,20 MICH J. INT'L L. 697, 714-19 (outlining a description of the investor arbitration regime). 5. Estimates are that over 3000 BITs were signed between 1959 and 2009. Jes- wald Salacuse, The Emerging Global Regime for Investment, 51 HARV. INT'L L.J. 427, 429 (2010). 6. Here, we are referencing the history and intent of BITs, not more recent jurisprudence in which the argument has been made that foreign investors actually are treated better than domestic investors, due to the right of recourse made available by BITs. 7. See U. N. CONFERENCE ON TRADE AND DEV., INVESTOR-STATE DISPUTES: PRE- VENTION AND ALTERNATIVES TO ARBITRATION, 3, U.N. Doc. UNCTAD/DIAE/IA/2009/11, U.N. Sales No. E.10.II.D.11 (2010) available at http://www.unctad.org/en/docs/ diaeia200911_en.pdf [hereinafter INVESTOR-STATE DISPUTES],; Ibrahim F.I. Shihata, Towards a Greater Depoliticization of Investment Disputes: The Roles of ICSID and MIGA, 1 ICSID REV. 1, n. 1 (1986), cited in Jack J. Coe, Jr., Transparency in the Resolution of Investor-State Disputes - Adoption, Adaptation, and NAFTA Leader- ship, 54 KAN. L. REV. 1339, 1345, n. 32 (2006) [hereinafter Transparency]. Spring 2013] Bilateral Investment Treaty Arbitration to resolve disputes and maintain relationships; (2) providing appropriate remedies to harmed investors; and (3) attracting foreign investment to those states that adhere to the investment treaty regime.8 Sometimes, these positive effects have been direct, as evi- denced by the issuance of arbitral awards followed by parties' peace- ful compliance. More frequently, arbitration's effects have been indirect, with states and investors communicating and negotiating toward resolution in order to achieve shared interests and at the same time avoid the arbitral forum.9 As W. Michael Reisman has observed recently, only a "miniscule fraction of the universe of foreign direct investment"10 finds its way to investment treaty arbitration. Nonetheless, the process can cast a long shadow."

8. See INVESTOR-STATE DISPUTES, supra note 7 at 3 ("Host states wishing to attract and promote foreign investment often seek to offer predictability to foreign investors by favouring international arbitration as the means for investors to deal with a dispute."). But see U. N. CONFERENCE ON TRADE AND DEV., THE ROLE OF INTERNA- TIONAL INVESTMENT AGREEMENTS IN ATTRACTING FOREIGN DIRECT INVESTMENT IN DE- VELOPING COUNTRIES, xi, U.N. Doc. UNCTAD/DIAE/IA/2009/5, U.N. Sales No. E.09.II.D.20 (2009) ("IIAs are part of the policy framework for foreign investment... HAs alone can never be a sufficient policy instrument to attract FDI [foreign direct investment]"), available at http://www.unctad.org/en/docs/diaeia20095_en.pdf; Jason Yackee, Do Bilateral Investment Treaties Promote Foreign Direct Investment? Some Hints from Alternative Evidence, 51 VA. J. INT'L L. 397, 438 (2011) (reporting research suggesting that a nation's entry into a BIT does not tend to influence companies' decisions to invest); Susan Rose-Ackerman & Jennifer L. Tobin, Do BITs Benefit Develop- ing Countries?, in THE FUTURE OF INVESTMENT ARBITRATION 134-36 (Catherine A. Rogers & Roger P. Alford, eds., 2009) (concluding that countries with poor investment environments do not benefit significantly from entering into BITs); Susan D. Franck, Empirically Evaluating Claims About Investment Treaty Arbitration, 86 N.C. L. REV. 1, 13-23 (2007) (surveying empirical research regarding investment arbitration). 9. See Jack J. Coe, Jr., Toward a Complementary Use of Conciliation in Inves- tor-State Disputes-A Preliminary Sketch, 12 U.C. DAVIS J. INT'L L. & POL'Y 7, 35 (2005) [hereinafter Complementary Use] (estimating that 30% of all arbitral filings are resolved by settlement); Jeswald W. Salacuse, Is There A Better Way? Alternative Methods of Treaty Based Investor-State Dispute Resolution, 31 FORDHAM INT'L L.J. 138, 168 (2007) (describing how "initiation of an arbitration may both favor and dis- courage negotiated settlements"); INT'L CTR. FOR SETTLEMENT OF INV. DISPUTES, THE ICSID CASELOAD STATISTICS 13 (2011), available at http://icsid.worldbank.org/ICSID/ Index.jsp (follow "Publications" hyperlink; then follow "ICSID Publications" hyperlink; then follow "The ICSID Caseload - Statistics" hyperlink; then choose from "Issue 2011-1") (indicating that 39% of ICSID's arbitration and conciliation proceedings that concluded in the first half of 2011 were settled or otherwise discontinued). 10. See W. Michael Reisman, International Investment Arbitration and ADR: Married but Best Living Apart, in INVESTOR-STATE DISPUTES II, supra note 1, at 22-27. 11. See Robert H. Mnookin & Lewis Kornhauser, Bargainingin the Shadow of the Law: The Case of Divorce, 88 YALE L.J. 950 (1979); see also Salacuse, supra note 9, at 157 (adopting definition of ADR as "alternatives to both international arbitration and adjudication in domestic courts"); see also Andrea K. Schneider, Bargainingin the HarvardNegotiation Law Review [Vol. 18:71

Within the last decade, however, there has been a noticeable surge in the number of investors accessing investment treaty arbitration. 12 Perhaps as a result of this phenomenon, 13 coupled with the heavy costs of investment treaty arbitration 14 and the magnitude of some arbitral awards, both states and investors are now raising multiple concerns. They decry the expense, delays, and political challenges associated with relying exclusively on a rights-based arbitral process and its outcomes.' 5 The current investment arbitration system relies on states' voluntary compliance with arbitral awards, especially when they are adverse. Some states, however, have threatened to refuse to abide by arbitral awards, others have acted on such threats, and still others have chosen to withdraw from BITs.' 6 In a few states, popular resistance to the prospect of investor-state arbitration has manifested in rioting.' 7 Such opposition and the implementation of threats of non-compliance have the potential to

Shadow of (International)Law: What the Normalizationof Adjudication in Interna- tional Governance Regimes Means for Dispute Resolution, 41 N.Y.U. J. INT'L L. & POL. 789 (2009). 12. See INT'L CTR. FOR SETTLEMENT OF INV. DISPUTES, supra note 9; Salacuse, supra note 9, at 143-47. 13. Significant surges in other contexts have led to similar perceptions of crisis and the need for change. See e.g., Marc Galanter, Real World Torts: An Antidote to Anecdote, 55 MD. L. REV. 1093, 1107 (1996) (noting that surge in asbestos cases may have contributed to a sense that courts were overburdened); Scott Sigmund Gartner & Gary M. Segura, War, Casualties,and Public Opinion, 42 J. CONFLICT RESOL. 278, 296-99 (1998) (observing that sudden surges in key indicators-e.g., war deaths- cause change in institutional strategies). 14. See Susan D. Franck, Rationalizing Costs in Investment Treaty Arbitration, 88 WASH. U. L. REV. 769, 782-90 (2011) (reporting regarding the costs of investment treaty arbitration); Catherine Rogers, The Arrival of the "Have-Nots" in International Arbitration, 8 NEV. L. J. 341, 357 (2007) (observing that while foreign investors have typically hired major international law firms to represent them in investor-state arbitration, many developing countries have not, due to the expense associated with such representation or for political reasons). 15. See Thomas J. Stipanowich, Arbitration: The "New Litigation," 2010 U. ILL. L. REV. 1, 29 (2010); JAMES M GAITIS, COLLEGE OF COMMERCIAL ARBITRATORS GUIDE TO BEST PRACTICES IN COMMERCIAL ARBITRATION 2-5 (2d ed. 2010); Salacuse, supra note 9. 16. INVESTOR-STATE DISPUTES, supra note 7, at 15-16; Franck, supra note 8, at 64-65. See also Issues Note, United Nations Conference on Trade and Development, Denunciation of the ICSID Convention and BITs: Impact on Investor-State Claims, U.N. Issues Note No. 2, at 4-5 (Dec. 2010) (describing withdrawal of Bolivia and Ecua- dor from ICSID and Ecuador's termination of several bilateral investment treaties), available at http://www.unctad.orglen/docs/webdiaeia20lO6_en.pdf; Ignacio A. Vincentelli, The Uncertain Future of ICSID In Latin America, 16 LAW. & Bus. REV. AM. 409 (2010) (reviewing the growing resistance in Latin American to ICSID). 17. Ben Lewis, Arbitration Targeted in FinalFight Over Korea FTA Passage,THE ASIAN LAWYER (Nov. 18, 2011), http://www.law.com/jsp/law/international/LawArticle- Intl.jsp?id=1202532852271&slreturn=1. The strength of this opposition is particularly noteworthy in light of empirical research showing that investment arbitral Spring 20131 Bilateral Investment Treaty Arbitration inject substantial uncertainty into the current system of international investment and trade, especially if the international community fails to find some means to buttress the legitimacy of investment treaty arbitration and its awards.' 8 Integrating mediation into the process choices offered by the current system represents one important response that could and should be made by the international community. Mediation represents an attractive option in this context for several reasons. First and most obviously, the process provides investors and states with the potential to resolve their disputes themselves-more quickly than arbitration, less expensively than arbitration, and in a manner designed to preserve a valuable business relationship. Indeed, mediation even has the potential to improve the business relationship if investors and states use the process to share important information, including their most important needs or interests, and take advantage of the opportunity to build trust in each other, or at least reduce distrust. Finally, mediation offers the opportunity to include other parties, who may not have standing to participate in the arbitration process, but whose participation may expand available resources and assist with the implementation of any agreements that are reached. This Article begins with a brief description of some of the problems that have arisen in the current investment treaty system, using the experience in Argentina as an example. The Article turns next to mediation and the concerns that have been raised regarding its potential integration into the investor-state context. Responding to these concerns requires consideration of dispute system design principles, research regarding the antecedents and influence of perceptions of procedural justice, and theories explaining such influence. The Article therefore goes into some detail regarding dispute system

tribunals have not tended to favor one side over the other. See Tai Heng Cheng, Inter- national Mediation, Arbitration and Innovation, in CONTEMPORARY ISSUES IN INTER- NATIONAL ARBITRATION AND MEDIATION: THE FoRDHAM PAPERS 2010 434, 438 (Arthur W. Rovine ed., 2010) [hereinafter CONTEMPORARY ISSUES 2010] (citing Susan D. Franck, Development and Outcomes of Investment Treaty Arbitration, 50 HARV. INT'L L.J. 435, 439 (2009)). 18. See, e.g., ERNST-ULRICH PETERSMANN, THE GATT/WTO DISPUTE SETTLEMENT SYSTEM xIv-xv, 3-4 (1997) (regarding non-compliance with arbitral awards arising out of the GATT and subsequent replacement with WTO system). Threats of non-compliance certainly are not new, or even unique to international arbitration. In the U.S., the famous case of Marbury v. Madison involved a freshly-minted Supreme Court similarly struggling to establish its legal and political authority while also acknowl- edging its dependence on the enforcement power wielded by the executive branch of government. See Marbury v. Madison, 5 U.S. 137 (1803). Harvard Negotiation Law Review [Vol. 18:7 1 design and procedural justice. The Article then turns to the relationship between mediation, on one hand, and the processes of arbitration, conciliation, and preventative procedures adopted by some states, on the other hand. This examination requires consideration of the different models of mediation and their advantages and disad- vantages. The Article then reports on the variety of compulsory referral schemes that have been tested in the U.S. and elsewhere and finally touches briefly on the importance and variety of available quality control measures and the development of a pool of investor- state mediators. Ultimately, and based on dispute system design principles and procedural justice research and theories, this Article will recommend the integration of a "default" model of mediation into the investor- state context. This model begins in a facilitative manner, in order to increase the likelihood of trust-building and information exchange, but also permits evaluative interventions and the use of legal norms. The Article further concludes that if stakeholders' input is sought and considered regarding mechanisms for the referral of disputes to mediation, some elements of mediation could be made compulsory. Particularly, dispute resolution clauses in investment treaties could require the parties' participation in an initial meeting, or even an initial mediation session, with the parties themselves then choosing whether to proceed further in mediation at that time or to specify the timing of its future use. Finally, the Article recommends the establishment of a small pool of well-known and well-respected investment treaty mediators who will offer a reasonably strong and pragmatic guarantee of quality in the short-term and engender a heightened perception of trust in the process. These mediators should also possess the temperament and skills to provide the default model of mediation. In the long term, evaluation and mentoring must be put into place to permit thoughtful cultivation of both the model of mediation that is best suited for the investor-state context and the next generation of mediators.

II. THE CURRENT STATE OF INVESTOR-STATE ARBITRATION AND CONCERNS ABOUT MEDIATION A. The Emergence of Problems in Investor-State Arbitration Despite the notable successes of investment treaty arbitration, described supra, investors and states are now raising multiple concerns. Of course, the significant costs of the process and the magnitude of some of the awards explain many of these concerns. But the current arbitration process also tends to focus parties on the law and their legal rights-and the need to resolve ambiguities regarding Spring 20131 Bilateral Investment Treaty Arbitration such legal rights' 9-when the parties' extra-legal interests may be both equally important and sufficient to achieve a meaningful, last- ing resolution. 20 Indeed, some commentators fear that the win-lose nature of the arbitration process itself may serve to marginalize some parties' uniquely cooperative socio-cultural characteristics and in- hibit them from identifying the mutual interests that have the potential to keep a troubled business relationship from becoming irreparably broken. As Jeswald Salacuse notes, "[n]either the aim nor the consequence of arbitration is to repair a broken business relationship." 21 Particularly noteworthy have been the comments made by Grant Kesler, the Chief Executive Officer of Metalclad, after his company won a 17-million-dollar arbitral award against Mexico. In retrospect, and despite being victorious, Kesler said that arbitration had been so dissatisfying that he wished his company had relied upon its "political options" to resolve the dispute. 22 Gabriel Bottini, who leads the International Affairs Department in the Solicitor-General's Office for Argentina, has similarly cited his own experience to opine that "at least for some" of the investors involved in disputes with Ar- gentina, "conciliation could have been a valuable alternative to pursue the resolution of these claims. '23 These comments suggest that

19. See Mark A. Clodfelter, Why Aren't More Investor-State Treaty Disputes Set- tled Amicably?, in INVESTOR-STATE DISPUTES II, supra note 1, at 38-42 (citing to legal questions regarding the arbitrability of disputes arising under most-favored-nation clauses, the types of activities that qualify as investments, and the elements of claims for violation of the fair and equitable treatment standard or the national treatment standard). 20. See Jennifer Gerarda Brown, Peacemaking in the Culture War Between Gay Rights and Religious Liberty, 95 IowA L. REV. 747 (2010) (suggesting that mediation is most appropriate to craft customized implementation of judicially-determined legal entitlements). 21. Salacuse, supra note 9, at 155. 22. See Susan D. Franck, Integrating Investment Treaty Conflict and Dispute Systems Design, 92 MINN.L. REV. 161, 224 n. 266 (2007) (citing Coe, Complementary Use, supra note 9, at 8-9); Salacuse, supra note 9, at 147; see also Jack J. Coe, Jr., Should Mediation of Investment Disputes Be Encouraged, and, if so, by Whom and How?, CONTEMPORARY ISSUES IN INTERNATIONAL ARBITRATION AND MEDIATION: THE FORDHAM PAPERS 2009, at 339, 339-40 [hereinafter CONTEMPORARY ISSUES 2009] (observing that Metalclad received $7 million less than it sought and found its hopes of building upon its relationship with Mexico dashed). 23. Gabriel Bottini & Veronica Lavista, Conciliationand BITs, in CONTEMPORARY ISSUES 2009, supra note 22 at 359, 365 (Arthur W. Rovine ed., 2009). But see Peter M. Wolrich, Resolution of Disputes by ICC Dispute Boards, in CONTEMPORARY ISSUES 2010, supra note 17, at 475 (describing the differences among the ICC's Dispute Re- view Boards, Dispute Adjudication Boards and Combined Board, and observing that debtors tend to prefer the review process-which allows them to retain control of the monies in dispute unless they are persuaded to pay the creditor-while creditors tend HarvardNegotiation Law Review [Vol. 18:71 losing parties, and even some of those who have won, may not perceive arbitration as providing them with a sufficient opportunity to come to their own less public but more customized, satisfying and final resolutions. The series of cases arising out of Argentina's debt crisis early in this century provide concrete examples of arbitration's potential dis- advantages. 24 In the early1990s, many foreign corporations invested in Argentinian companies as Argentina promoted privatization. As part of these investments, Argentina agreed to stabilize the peso against the dollar by collecting tariffs in dollars and readjusting the tariff rate twice a year. (U.S. investors benefitted economically from this assurance of a stable tariff.) Ten years later, during the Argen- tinian debt crisis, the government of Argentina passed an emergency law suspending both the favorable conversion ratio and the semi-annual adjustments. (Because these stabilization programs had only applied to foreign investors, the suspension disadvantaged only the foreign investors.) When foreign companies brought arbitration cases under the BIT, claiming that foreign investors had been denied fair and equitable treatment, Argentina argued the necessity defense under customary international law and the relevant BIT's emergency clauses. One such arbitration involved CMS Energy, an American corpo- ration that owned 30% of TGN, an Argentinean gas transportation company. As part of CMS's original investment, the government of Argentina agreed to the favorable conversion ratio and semi-annual adjustments described supra.25 At arbitration, CMS won an award of $133 million against Argentina in 2005.26 Argentina moved to annul the award, and CMS won again in 2007 before the annulment committee. 27 The annulment committee, however, so criticized the legal to prefer the adjudicative process-which may result in the debtor being forced to pay the creditor). 24. See Andrea Kupfer Schneider, Error Correctionand Dispute System Design in Investor-State Arbitration, 5 PENN ST. Y.B. ON ARB. & MEDIATION (forthcoming 2013) (describing the series of cases in greater detail). See also Irene Ten Cate, Interna- tional Arbitration and the Ends of Appellate Review, 44 N.Y.U. J. INT'L L. & POL. 1109, 1174 (2012). See also David Schneiderman, JudicialPolitics and International Investment Arbitration: Seeking An Explanation for Conflicting Outcomes, 30 Nw. J. INT'L L. & Bus. 383 (2010). 25. See Cate, supra note 24. 26. CMS Gas Transmission Co. v. Argentine Republic, ICSID Case No. ARB/0118, Award, 468 (May 12, 2005) https://icsid.worldbank.org/ICSID/FrontServlet?request Type=CasesRH&actionVal=showDoc&docId=DC504_En&caseId=C4. 27. CMS Gas Transmission Co. v. Argentine Republic, ICSID Case No. ARB/01/8, Annulment Decision, 159, 163 (Sept 25, 2007) https://icsid.worldbank.orgICSID/ FrontServlet?requestType=CasesRH&actionVal=showDoc&docId=DC505_En&case Spring 2013] Bilateral Investment Treaty Arbitration basis for the original arbitral award that Argentina likely perceived itself as quite justified in not paying.28 Today, seven years after the original award, CMS has yet to receive payment from Argentina. Rather, CMS has transferred the debt to a subsidiary of Bank of America that specializes in distressed debt, to permit CMS to continue to invest in Argentina. 29 Other significant awards against Ar- gentina also remain unpaid. The award won by the U.S. company 30 Azurix, for example, amounts to more than $165 million. Argentina has successfully used the annulment process in other cases, however. Both the Sempra and Enron arbitral tribunals, for example, used reasoning similar to that of the original CMS tribunal, to find for the investor-claimants and against Argentina. 31 When Argentina moved to annul these awards, however, it was successful. 32 Argentina has also achieved partial annulments to reduce arbitral awards and has settled other cases before they have gone to an 33 arbitration hearing.

Id=C4. Under ICSID, foreign investors file a request for arbitration with the Secre- tary-General of ICSID. ICSID art. 36. Cases proceed to arbitration under ICSID rules heard by an arbitral tribunal. If a dispute arises as to the award, parties can request an annulment of the award. This annulment committee is constituted, also by the Secretary-General of ICSID, of three new arbitrators to review the award. IC- SID art. 52. The selection of annulment committee members and the annulment committee process is worthy of additional academic study but is beyond the scope of this article. 28. See David Caron, Framing the Work of ICSID Annulment Committees, 6 WORLD ARB. & MEDIATION REV. 173 (2012), available at http://works.bepress.com/ davidcaron/134/ (quoting Jose Alvarez in fn 18). 29. Luke Eric Peterson, Argentine Crisis ArbitrationAwards Pile Up, but Inves- tors Still Wait for a Payout, LAw.COM, June 25, 2009, http://www.law.comjsp/article. jsp?id=120243173673 1&Argentine CrisisArbitrationAwards_Pile_UpbutInvest orsStillWait for a Payout (last visited Oct. 28, 2012). 30. Azurix v. Argentine Republic, ICSID Case No. ARB/01/12, Annulment Deci- sion, $137 (Sept. 1, 2009) https://icsid.worldbank.org/ICSID/FrontServlet?requestType =CasesRH&actionVal=showDoc&docId=DC1171_En&caseId=C5. (upholding the arbitral award). 31. Sempra Energy Int'l v. Argentine Republic, ICSID Case No. ARB/02/16, Award, 482(Sept. 28, 2007) http://italaw.com/documents/SempraAward.pdf; Enron Corp. v. Argentine Republic, ICSID Case No. ARB/01/3, Award, 91 261 (May 22, 2007), http://www.italaw.com/sites/default/files/case-documents/ita0293.pdf. 32. Sempra Energy Int'l v. Argentine Republic, ICSID Case NO. ARB/02/16, An- nulment Decision, 223 (June 29, 2010) https://icsid.worldbank.orgfICSID/FrontSer- vlet?requestType=CasesRH&actionVal=showDoc&docId=DC1550_En&cased=C8 ; Enron Corp. v. Argentine Republic, ICSID Case No. ARB/01/3, Annulment Decision, $[ 406-17 (July 30, 2010) http://italaw.com/documentsfEnronAnnulmentDecision.pdf. See Schneider, supra note 24, for more on the comparisons between the legal reasoning of the contradictory annulment committees decisions and the impact of this inconsistency. 33. See Lucy Reed, Scorecard of Investment Treaty Cases Against Argentina Since 2001, KLUWER ARBITRATION BLOG, Mar. 2, 2009, http://kluwerarbitrationblog. Harvard Negotiation Law Review [Vol. 18:71

In all of these cases, Argentina has claimed that its actions were the result of necessary economic policy decisions. Some companies, like CMS, have moved on and written off whatever they might have collected from the arbitration awards. Other companies have negotiated settlements with Argentina. Still others, however, continue to fight and have involved the U.S. government, which has now sus- pended Argentina from general preferential trade treaties. 34 Both the lack of resolution for individual companies and worsening economic relationships between governments are exactly the sorts of problems that the arbitration system was designed to avoid.

B. Adding Mediation to the Mix Responding to these concerns, Professors Jack Coe,35 Jeswald Salacuse, 36 and Susan Franck 37 have urged consideration of procedures and systems that actually could prevent the development of investor-state disputes. They have also recommended early and greater use of consensual procedures such as conciliation, early neutral evaluation, ombudspersons, and mediation. Some states have begun to put preventative procedures into place. 38 These will be described in greater detail infra. Investors, states, and interested international bodies and scholars have also begun to focus quite specifically on mediation. 39 As noted supra, mediation is attractive because it provides investors and states with the potential to resolve their disputes themselves-more quickly than arbitration, less expensively than arbitration, and in a manner designed to preserve a valuable business relationship. The process also has the potential to encourage investors and states to develop trust in each other (or at least reduce distrust) and share important information, including their most important needs or interests that must be met in order to arrive at a resolution. Mediation's flexibility also offers the opportunity to com/blog/2009/03/02/scorecard-of-investment-treaty-cases-against-argentina-since- 2001/, (last visited Oct. 28, 2012). 34. Tom Barkley & Ken Parks, U. S. Cuts Trade Preferences to Argentina, WALL ST. J., Mar. 26, 2012, http://online.wsj.com/article/SB100014240527023041771045773 05652479085184.html (March 26, 2012). 35. See Coe, Complementary Use, supra note 9; see also Coe, supra note 22. 36. See Salacuse, supra note 9. 37. See Franck, supra note 14. 38. See infra Section III(a). 39. See, e.g., INVESTOR-STATE DIsPuTEs II, supra note 1; IBA Rules For Investor- State Mediation Int'l Bar Ass'n, Mediation Comm., State Mediation Subcomm., (Oct. 4, 2012) [hereinafter IBA Rules]. Spring 20131 Bilateral Investment Treaty Arbitration include other parties, who may not have standing to participate in the arbitration process, but whose participation may expand available resources and assist with the implementation of any agreements reached by the parties. Finally, the mediator can help both the state and the investor to be realistic about their options, thus making it more likely that the parties will arrive at a solution they can, and will, implement. The International Bar Association ("IBA") very recently ap- proved rules to facilitate the use of investor-state mediation (IBA Rules for Investor-State Mediation or "IBA Rules").40 This action by the IBA represents an extremely important first step toward legiti- mizing investor-state mediation. At the same time, the rules are unlikely to go far enough to motivate significant integration of mediation into the fabric of the investment treaty context. First, the IBA Rules provide only for voluntary mediation, beginning with a mediation management conference which occurs only after the parties have elected mediation and designated the mediator.41 As outlined infra in this Article, some form of compulsion may be needed, at least provisionally, to encourage parties to overcome current obstacles to the use of mediation.4 2 As this Article will explain, such compulsion need not be terribly intrusive. For example, investment treaties' dispute resolution clauses could require the signatory states' and investors' participation in an initial meeting to discuss the

40. Id. 41. Interestingly, the Rules provide that "[bly agreeing to mediate under these rules, a party undertakes to participate in the mediation management conference." Id. at art. 9.4. Participation in the mediation management conference apparently represents the extent of a party's commitment: "A party may withdraw from the mediation at any time after the mediation management conference. Prior to withdrawing from the mediation, a party must notify the other party or parties and the mediator of its intention to withdraw, preferably stating its reasons. Prior to a party's withdrawal from the mediation, the mediator shall hold a meeting with all parties in person, by telephone or by any other means of telecommunication." Id. 42. Some academics and stakeholders have responded with interest to suggestions that mediation should not just be encouraged, but made compulsory. See Lisa Blomgren Bingham, Opportunitiesfor Dispute Systems Design in INVESTMENT TREATY DispuTEs: Consensual Dispute Resolution at Varying Levels, in INVESTOR-STATE Dis- PuTEs II, supra note 1, at 33 (observing that the conference included discussion of mandatory mediation, but with an opt-out); Wolf von Kumberg, Making Mediation Mainstream: An Application for Investment Treaty Disputes, in INVESTOR-STATE Dis- PUTES II, supra note 1, at 71 (proposing inclusion of mediation in a multi-step dispute resolution clause in BITs). But see Margrete Stevens, Synopsis of Remarks, in INVEs- TOR-STATE DispuTEs II, supra note 1, at 28 (not supportive of compulsory mediation); Lucy Reed, Synopsis of Closing Remarks, in INVESTOR-STATE DIsPUTES II, supra note 1, at 30 (not supportive of compulsory mediation). HarvardNegotiation Law Review [Vol. 18:71 potential use of mediation or other consensual procedures, before or soon after the initiation of arbitration proceedings. Second, the IBA Rules are silent regarding the "default" model of mediation that should be used to achieve the goals that motivated states to enter into BITs. The IBA Rules provide only that "[i]n conducting the mediation ... the mediator shall take into account the wishes of the parties. . .,43 Such failure to identify the "default" model of mediation makes it difficult to know what "mediation" is and how to differentiate the process from other available options, such as conciliation. If there is no such differentiation, it is unclear why mediation is needed. Finally, the IBA Rules may not be clear enough in providing that justice-particularly procedural justice-represents a central goal of the mediation process. The Rules provide that the mediator "shall be guided" by several principles, including the principle of "fairness," but later provide that in managing the process, the mediator "shall take into account" the parties' wishes, "the circumstances of the case and the overall goal of a cost-efficient and timely settlement of the differences or disputes."44 In failing to re-assert the role of procedural justice here, the IBA Rules could inadvertently undermine the legitimacy of mediation and its outcomes. Despite all of the concerns just raised, the IBA Rules definitely demonstrate the existence of an appetite for mediation as a new process choice and the commitment of stakeholders to move forward with its integration.

C. Critiques of Mediation-andthe Challenges to be Met

Not all stakeholders, however, have been so ready to embrace expanded use of mediation. First, some have criticized the mediation process as, at best, unnecessary and, at worst, a threat to the useful work being done by arbitration.45 Negotiation and conciliation

43. IBA Rules, supra note 39, at art. 7; see also id. at art. 8 (providing that "[tihe mediation shall be conducted in accordance with the parties' wishes . . . "). 44. Id. at 5 (emphasis added); Bobbi McAdoo & Nancy A. Welsh, Court-Connected General Civil ADR Programs: Aiming for Institutionalization,Efficient Resolution, and the Experience of Justice, in ADR HANDBOOK FOR JUDGES 1, 2 (Donna Stienstra & Susan Yates eds., 2004) (emphasizing the importance of defining the goals of establishing a court-connected ADR program and providing guidance regarding all aspects of implementing such a program). 45. See, e.g., Reisman, supra note 10; see generally Coe, Complementary Use, supra note 9, at 17-18 (describing the weaknesses of conciliation that arise out of its dependence upon the parties' voluntary and good faith use of the process). Spring 2013] Bilateral Investment Treaty Arbitration already provide diplomatic options for investors and states, and conciliation is rarely used. Given the current fragility of the investment arbitration system, adding a new process could be destabilizing. Other commentators have observed, meanwhile, that mediation and conciliation are largely interchangeable, 46 which suggests that the introduction of mediation will not produce any more resolution than currently results from negotiation and conciliation. These points recall debates that occurred in the U.S. a couple of decades ago, when mediation proponents first urged the integration of mediation into 47 civil litigation. These points also reveal the importance of identifying a "default" model of mediation that is meaningfully different from the existing processes of parties' negotiations and conciliation. Further, they suggest the need to incorporate some compulsory elements, identify a pool of mediators who have the credibility, trustworthiness, knowledge, experience and skills that the parties do not already possess, and provide a means to assure the continued quality of those 48 mediators providing investor-state mediation.

46. But see Salacuse, supra note 9, at 166 (estimates of negotiated settlements in ICSID run as high as 30%); See Coe, Complementary Use, supra note 9, at 14. 47. Some judges and litigators resisted that call, confusing "mediation" with "meditation" and therapy, labeling mediation a "fad," and insisting that mediation would represent an unprincipled, emotion-driven process. Some of these criticisms persist. See e.g., THOMAS CARBONNEAU, CARBONNEAU'S ARBITRATION IN A NUTSHELL, 3rd ed. 362-368 (2012) (describing mediation as a "'talking therapy' model of dispute resolution" that is most appropriate for the resolution of child custody disputes, "elit- ist and paternalistic at its foundation" and "built upon fallacious and incomplete thinking"). Interestingly, such critics have been joined by others who now suggest that mediation represents nothing more than the unprincipled bargaining of the mar- ketplace. See e.g., Kakani v. Oracle Corp., 2007 WL 1793774, 154 Lab. Cas. P 35,310 (N.D.Cal., June 19, 2007) ("It is also no answer to say that a private mediator helped frame the proposal. Such a mediator is paid to help the immediate parties reach a deal. Mediators do not adjudicate the merits. They are masters in the art of what is negotiable. It matters little to the mediator whether a deal is collusive as long as a deal is reached. Such a mediator has no fiduciary duty to anyone, much less those not at the table. Plaintiffs' counsel has the fiduciary duty. It cannot be delegated to a private mediator."). 48. A further difficulty involves conflicting cultural assumptions regarding parties' ability to listen and change in response to changing circumstances. See Michael W. Morris & Michele J. Gelfand, Cultural Differences and Cognitive Dynamics: Ex- panding the Cognitive Perspective on Negotiation, in THE HANDBOOK OF NEGOTIATION AND CULTURE 45, 52-53 (Michele J. Gelfand & Jeanne M. Brett eds., 2004) (describing cultural variability in dispositionist bias or fundamental attribution theory, with East Asians being more likely than North Americans to perceive an individual's behavior as situationally-contingent and influenced by the group); Philip McConnaughay, Re- thinking the Role of Law and Contracts in East-West Commercial Relationships, 41 HarvardNegotiation Law Review [Vol. 18:71

Second, some commentators argue that the entire legal regime would have to be restructured if mediation were added. Hundreds of BITs would require renegotiation and the rules and procedures of IC- SID and other bodies would need to be revised. Such renegotiation and revision 49 has the potential to be both time-consuming and wasteful, especially if parties can easily avoid the mediation process or participate in a manner that makes resolution unlikely. Once again, this objection reveals the importance of designing the mediation process, and the mechanisms to support its use, in a manner that makes good faith participation and resolution more likely and also saves costs, time and investor-state relationships, in both the short term and the long term. This is a primary reason for this Article-to plumb past experience with the institutionalization of compulsory mediation, as well as dispute system design and procedural justice research and theories, to suggest means that will make it more likely that investor-state mediation achieves its goals. A third critique of mediation in the investor-state context is based on the unique difficulties that arise in disputes that involve a sovereign state. Some commentators worry that mediation, especially if it includes compulsory elements, cannot possibly be exported from the private commercial context into the investment treaty context. These commentators argue that in international commercial mediation, private parties can participate more openly and enter into "deals" more quickly because they do not need to grapple with national security issues, economic policy issues, and the prospect of domestic political accountability. 50 While a state can frame an adverse arbitral award as an unfortunate loss that nonetheless requires adherence due to treaty obligations, 51 the state's voluntary settlement that acknowledges the commission of a wrong and payment of a substantial sum to the investor has the potential to be a much harder

VA. J. INT'L L. 427, 428-40 (2000-2001) (contrasting Western and Eastern understand- ings of contract obligations). This much larger issue will not be addressed in this Article. 49. Coe, supra note 22, at 350-52 (discussing the necessary revised treaty language). 50. See Reed, supra note 42, at 30-32. 51. See Salacuse, supra note 9, at 168. In particular, note the story of the Egyp- tian Pyramids case in which the president himself turned down a settlement offer because of political pressures. Even though the offer to settle-for $10 million- would have made economic sense for the government of Egypt, the president knew that press coverage of the settlement would sharply criticize the government. Press coverage, however, of the continuing arbitration was virtually nonexistent. See AMAzu A. Asouzu, INTERNATIONAL COMMERCIAL ARBITRATION AND AFRICAN STATES: PRACTICE, PARTICIPATION AND INSTITUTIONAL DEVELOPMENT 315-17 (2001). Spring 2013] Bilateral Investment Treaty Arbitration political decision to sell to constituents. This barrier will be particularly difficult to surmount if the investor was harmed due to the state's implementation of a public policy or regulations that clearly responded to the needs of its own people. In the CMS case described supra, economists have argued that Argentina had no choice but to devalue its currency in order to prevent further economic crises.52 How could the government negotiate the payment of millions of dollars to a foreign company to compensate it (and its domestic partners) for its economic loss when Argentina's citizens and wholly-owned domestic companies clearly had lost so much more? In addition, the state's authority to negotiate a voluntary settlement is often unclear and requires traveling through several layers of bureaucracy. 53 Many states have established a clear chain of com- mand and a methodology to manage their participation in the arbitration process. This is particularly true for those states that have had the unfortunate but instructive experience of defending themselves against investors' claims.5 4 These states, their lawyers, and officials in agencies and local units of government understand the process of drafting the appropriate documents for arbitration proceedings, making the best arguments possible, marshaling the necessary evidence, etc. In contrast, the internal negotiations among state officials, agencies, and even local units of government to reach consensus on settlement authority and a settlement range might be significantly more difficult and time-consuming. Indeed, these internal negotiations may be substantially more complex than the official negotiations across the table. 55

52. Peterson, supra note 29 (quoting Joseph Stiglitz). 53. In fact, lower-level diplomats might need protection in order to reach settlement. Reportedly, in one case in Ecuador, the government official was actually crimi- nally indicted for settling a case. 54. See Rogers, supra note 14, at 357 (discussing states that do not have sufficient experience with investment arbitration); Eric Gottwald, Leveling the Playing Field: Is it Time for a Legal Assistance Center for Developing Nations in Investment Treaty Arbitration?,22 AM. U. INT'L L. REV. 237, 255 (2007). 55. See INVESTOR-STATE DispuTEs, supra note 7 at 69-74; INVESTOR-STATE Dis- PuTEs II, supra note 1 at 51; Andrea K. Schneider, Getting to NAFTA: A Review of InterpretingNAFTA by Frederick W. Mayer, 17 BERKELEY J. INT'L L. 330, 338 (1999) (discussing the difficulties of two-level diplomacy in trade and investment agreements). These internal negotiations also may involve discussion of whether a state's "judgment fund" is, or should continue to be, available to pay settlements, and if it is, under what circumstances. See e.g., Payment of Judgments and Compromise Settle- ments, 28 U.S.C. § 2414 (2011); Judgment Fund Transparency Act of 2013, H.R. 317, 113th Cong. (2013). HarvardNegotiation Law Review [Vol. 18:71

While it is essential to recognize the unique constraints faced by sovereign nations,56 it is just as essential to recognize that mediation and other consensual processes are used regularly to resolve domestic public policy, civil rights and major regulatory disputes that involve extraordinarily difficult public and private issues and state actors. This is another primary reason for this Article. The proponents of investor-state mediation need to be aware of the successful use of mediation to resolve large, complex matters, as well as the relevance of dispute system design principles and procedural justice theories and research. Compliance with such principles, theories and research makes it more likely that investor-state mediation will produce outcomes that are perceived as fair and enhance the legitimacy of the governmental actors that are involved. A fourth, largely unspoken critique of mediation likely arises out of lawyers' fears of losing some measure of their professional autonomy.57 The current system of investment treaty arbitration clearly favors a traditional allocation of responsibilities between American lawyers and their clients. Under these rules, a state instructs its lawyers that its objective is to avoid making any payment to the foreign investor; the investor, meanwhile, instructs its lawyers that its objective is to win compensation for past and future losses. The lawyers then assume control over the "means" to achieve these relatively clear objectives in the context of arbitration.58 Sophisticated state officials and investor representatives are likely to review legal briefs and arguments, but even they are equally likely to delegate the authority to advocate. 59 Granting affirmative authority to settle, however, is a more nuanced matter. States might want to retain more

56. Negotiations with sovereign states inevitably can bring up issues of sovereign immunity, both in the domestic and foreign court system. Since the investor-state treaty system relies on a state's waiver of this immunity (at least for the purposes of arbitration), this article will not address that issue. Further academic study as to how and when states should negotiate away their immunity is warranted. 57. See Nancy A. Welsh, Looking Down the Road Less Traveled: Challenges to Persuading the Legal Profession to Define Problems More Humanistically, 2008 J. Disp. RESOL. 45, 51-53 (2008) (exploring how the use of factors other than legal or litigation-related analysis could threaten law's claim to significance and lawyers' professional autonomy and authority); John Lande, Escaping from Lawyers' Prison of Fear (describing events that could lead to loss of autonomy) (manuscript on file with author). 58. See MODEL RULES OF PROF'L CONDUCT R. 1.2 (2002). Note that the Rule provides that lawyers are to consult with their clients regarding the means to be used to achieve the clients' objectives. 59. Increasingly, however, practitioners and commentators are encouraging clients and inside counsel to manage outside counsel more actively. See Sigfried H. Elsing, On Babies and Bathwater: Keeping the Good (And Getting Rid of the Bad) Spring 2013] Bilateral Investment Treaty Arbitration control or might worry that their lawyers have insufficient understanding of the political and business consequences of settlement. If a state wishes to pursue settlement and arbitration simultane- ously,60 it must establish the delegation of additional responsibilities and authority for settlement. Lawyers representing the states will also have to become familiar, if they are not already, with mediation representation skills. These skills-e.g., exploring underlying interests as well as the facts that are relevant to legal arguments, writing mediation submissions, preparing clients for mediation, explaining clients' positions persuasively to the other side, considering how the mediator may assist communication, information-gathering, and negotiation-are all related to traditional advocacy skills, 61 but also differ because they assume that the ultimate audience is the other disputant, rather than an arbitrator or panel of arbitrators. The challenge of learning and implementing these new perspectives and skills may help to explain American law firms' hesitation to embrace mediation. Their expertise and approach have been uniquely suited to domination of the legal frontier that is investor- state arbitration, 62 and they may not relish the need to adopt a more nuanced approach or add lawyers (sometimes called "settlement counsel") with expertise in convening and consensus-building and a

From the Company's Perspective, in CONTEMPORARY ISSUES 2010, supra note 17, at 307, 318-19 (discussing the need for parties to manage outside counsel in international arbitration processes, but also noting the lack of a legal professional privilege extending to in-house counsel); Frank H. Minaker et al., Arbitration of International Commercial Disputes, in CONTEMPORARY IssuEs 2010, supra note 17, at 322, 327-32 (discussion of roles of outside counsel, inside counsel and business persons in arbitrat- ing a commercial matter); see also Rogers, supra note 14. 60. Salacuse, supra note 9, at 166 (citing to an investor that stayed consistently open to negotiation even after commencing arbitration). 61. See David H. Burt, Inside Counsel as Sophisticated Users of the Mediation Process, in CONTEMPORARY IssuEs 2010, supra note 17, at 418, 419-28 (discussion of the various important roles that inside counsel can and should play in preparing for commercial mediation, including conducting early case assessment, developing a case narrative, managing outside counsel, interacting with inside counsel of the opposing party, and generally acting as process manager, intermediary and interpreter, skep- tic, evaluator and negotiator); see also David Wilkins, Team of Rivals? Toward a New Model of the CorporateAttorney-Client Relationship, 78 FORDHAM L. REV. 2067, 2091- 03 (2010) (describing integration and knowledge transfer between international law firms and multinational corporations, thus suggesting a growing convergence between outside counsel and their corporate clients); see generally Hal Abramson, MEDI- ATION REPRESENTATION: ADVOCATING AS A PROBLEM-SOLVER (3rd ed., 2013). 62. See Welsh, Looking Down the Road Less Traveled, supra note 57 at 57-59 (discussing psychological, professional and business reasons that lawyers may resist mediation and other humanistic approaches to resolving disputes); Roger P. Alford, The American Influence on InternationalArbitration, 19 OHIO ST. J. ON Disp. RESOL. 69, 80-82 (2003). Harvard Negotiation Law Review [Vol. 18:71 cooperative style.63 States and investors, meanwhile, may find it easier to delegate the authority to advocate than delegate the authority required to fashion a complex commercial settlement that also recog- nizes political realities. Several countervailing factors, however, suggest that the fears just identified are unlikely to be effective in inhibiting the integration of mediation. Probably most important is the fact that most of these disputes settle already, albeit without conciliation, and the lawyers in this context, as in so many others, are directly involved in such settlement. In addition, based on domestic and international experience, many investors and states are already familiar with a court- connected or court-oriented legal process called "mediation."64 Law- yers play a central role in this process, whether it is used in bilateral disputes, complex commercial matters, or class actions. Increasingly, investors, states and lawyers also have experience with the somewhat different forms of mediation and facilitation used to resolve large public policy disputes, including environmental issues involving many stakeholders.6 5 Many states also generally welcome the use of consensual processes as part of their cultural heritage.

63. See Bobbi McAdoo & Nancy A. Welsh, Look Before You Leap and Keep On Looking: Lessons from the Institutionalizationof Court-Connected Mediation, 5 NEV. L.J. 399 (2005) (based on the history of court-connected mediation in the U.S., ques- tioning the likelihood that lawyers will accept and promote processes designed to enhance parties' self-determination, consider their extra-legal interests and offer the potential for creative, mutually-beneficial solutions); but see Nancy A. Welsh, The ReputationalAdvantages of DemonstratingTrustworthiness: Using the Reputation In- dex with Law Students, 28 NEGOT. J. 117, 130-34 (2012) (discussing the value to lawyers of having a cooperative style) [hereinafter Reputational Advantages]; GERALD WILLIAMS & CHARLES CRAVER, LEGAL NEGOTIATING 53-54 (2007); Charles Craver, Ne- gotiation Styles: The Impact on BargainingTransactions, 49 J. Disp. RESOL. 48, 50-53 (2003). 64. See Suzanne Ulicny, ICC ADR: Rules and Approaches for Reaching an Amica- ble Solution, in CONTEMPORARY ISSUES 2010, supra note 17, at 478, 478-80 (reporting that ICC offers mediation, neutral evaluation, and minitrials, but that mediation is the most popular of these processes); Luis M. Martinez & Thomas Ventrone, The InternationalCentre for Dispute Resolution Mediation Practice, in CONTEMPORARY IS- SUES 2010, supra note 17, at 484, 484 (reporting that mediation is the process most frequently used, although ICDR also offers early neutral evaluation, fact-finding, minitrials, dispute resolution boards, partnering, double-blind bid settlement, and non-binding arbitration). 65. See e.g., Barbara Gray & Julia Wondolleck, Environmental Disputes: Negoti- ating Over Risks, Values and the Future, in HANDBOOK OF RESEARCH ON NEGOTIATION (Mara Olekalns & Wendi Adair, eds., forthcoming 2013), Sean F. Nolon, Negotiating the Wind: A Framework to Engage Citizens in Siting Wind Turbines, 12 CARDOZo J. CONFLICT RESOL. 327 (2011); Lawrence E. Susskind, Consensus Building and ADR: Why They Are Not the Same Thing, in THE HANDBOOK OF DISPUTE RESOLUTION (Michael L. Moffitt & Robert C. Bordone, eds., 2005) [hereinafter HANDBOOK OF Dis- PUTE RESOLUTION]; THE PROMISE AND PERFORMANCE OF ENVIRONMENTAL CONFLICT Spring 2013] Bilateral Investment Treaty Arbitration

Whether states are embracing mediation by developing their own corps of quasi-mediators or bringing investment arbitration to a point of crisis by withdrawing from BITs, the stage is set for the integration of mediation into the investment treaty context.66 The Inter- national Bar Association's recent approval of rules to facilitate the use of investor-state mediation offers substantial evidence that we are reaching the "tipping point."67 These developments also suggest the need for discipline and precision in defining the model or models of mediation that will be used, the breadth of any compulsory elements, mechanisms for providing transparency and ensuring quality, and the identity and role of the mediators. Such discipline and precision will come from adherence to the principles of dispute system design and the research and theories of procedural justice.

III. DISPUTE SYSTEM DESIGN AND PROCEDURAL JUSTICE A. Dispute System Design No dispute or dispute resolution process exists in a vacuum. Rather, every "conflict, issue, dispute, or case submitted to any institution for managing conflict (including one labeled ADR [alternative dispute resolution]) exists in the context of a system of rules, processes, steps, and forums. In the field of ADR, this is called dispute system design. 68 Dispute system design is based on an amal- gam of conflict theory, theories of organizational development, and an understanding of both "traditional" and "alternative" dispute resolution. It provides guidance regarding the process to be used in struc- turing a system, determining the component parts of the system, and measuring the system's effectiveness.

RESOLUTION (Rosemary O'Leary & Lisa B. Bingham, eds., 2003); THE CONSENSUS BUILDING HANDBOOK: A COMPREHENSIVE GUIDE TO REACHING AGREEMENT (Lawrence Susskind et al., eds., 1999); Nancy A. Welsh & Barbara Gray, Searching for a Sense of Control: The Challenge Presented by Community Conflicts Over Concentrated Animal Feeding Operations, 10 PENN ST. ENVTL. L. REV. 295 (2002). 66. This balance between exit, voice, and loyalty to international regimes re- mains an interesting one to explore as we move forward in adjusting the investment treaty system. See ALBERT 0. HIRSCHMAN, EXIT, VOICE, AND LOYALTY: RESPONSES TO DECLINE IN FIRMS, ORGANIZATIONS, AND STATES (1970) (explaining this concept); Jo- seph H.H. Weiler, The Transformationof Europe, 100 YALE L.J. 2403 (1991) (applying this concept to the European Union). 67. See IBA Rules, supra note 39. Similarly, the balance between private and public international law in the mixed context of investor-state dispute resolution is also interesting. See Andrea Kupfer Schneider, Private and Public InternationalDis- pute Resolution, in THE HANDBOOK OF DISPUTE RESOLUTION, supra note 65. 68. Lisa Blomgren Bingham, Reflections on Designing Governance to Produce the Rule of Law, 2011 J. DIsP. RESOL. 67, 74 (2011). HarvardNegotiation Law Review [Vol. 18:71

Based on field experience resolving disputes in the coal industry, William Ury, Jeanne Brett and Stephen Goldberg first wrote about dispute system design in their 1988 book, Getting Disputes Re- solved.69 They found that disputes in the workplace often are resolved through the use of power and rights, rather than interests. When organizations focus on achieving power-based or rights-based solutions, they miss the opportunity to find better solutions, better engage their stakeholders, and save money.70 The second generation of dispute system design, captured in Cathy Costantino and Christina Sickles-Merchant's book, Designing Conflict Management Systems,71 discusses how organizations create ADR methods most responsive to their needs in advance of the ripening of conflict. 72 In thinking about the array of choices available to organizations, they outline six categories of ADR processes: (1) preventative (e.g., dispute

69. WILLIAM URY ET AL., GETTING DISPUTES RESOLVED: DESIGNING SYSTEMS TO CUT THE COSTS OF CONFLICT (1988). 70. Id. at 7-8. In their book, Ury, Brett, and Goldberg outline six key principles for designing a presumptively interests-oriented dispute resolution system: (1) put the focus on interests; (2) build in opportunities to return [or "loop-back"] to negotiations ("Loop-backs" are defined as the opportunity to continue to move around in the process choices. So, for example, the parties should be able to go back and negotiate at any stage outlined in a dispute resolution process and not be limited to a "negotiation" stage that occurs early on. Similarly, the term "loop forward," developed in later DSD literature, also means that parties can choose to jump around among the process choices choosing to engage, for example, in fact-finding before negotiation.) (3) provide low-cost rights and power backups to interest-based processes; (4) build in consultation before creating the dispute system and feedback after the implementation and use of the system; (5) arrange procedures in a low-to-high-cost sequence; and (6) provide the necessary motivation, skills, and resources to permit participants and the organization to begin with a focus on interests and then move to assertions of rights and power only as necessary. Id. at 42. This prescription assumes, of course, that an organization's goals include productivity and inexpensive, speedy resolution of conflicts. An organization with the goals of avoiding change or maintaining stability might not be attracted to dispute system design. 71. CATHY A. COSTANTINO & CHRISTINA SICKLES MERCHANT, DESIGNING CONFLICT MANAGEMENT SYSTEMS: A GUIDE TO CREATING PRODUCTIVE AND HEALTHY ORGANIZA- TIONS (1996). 72. Costantino and Merchant outline their principles as: (1) developing guidelines for whether ADR is appropriate; (2) tailoring the ADR process to the particular problem; (3) building in preventative methods of ADR; (4) making sure that disputants have the necessary knowledge and skill to choose and use ADR; (5) creating ADR systems that are simple to use and easy to access and that resolve disputes early, at the lowest organizational level, with the least bureaucracy; and (6) allowing disputants to retain maximum control over the choice of ADR method and selection of neutral wherever possible. Id. at 120-121. Though not the primary focus of this Arti- cle, it is also important to note that Costantino and Sickles-Merchant examine how new dispute systems are developed, observing that some organizational leaders had used rights-based mechanisms to impose interest-based processes upon stakeholders. Id. Spring 2013] Bilateral Investment Treaty Arbitration resolution clauses, partnering, consensus building), (2) negotiated; (3) facilitated (e.g., mediation, conciliation, institutional ombuds); (4) fact-finding (e.g., neutral experts, masters); (5) advisory (e.g., early neutral evaluation, non-binding arbitration);73 or (6) imposed (e.g., 74 binding arbitration). Applying this framework to investor-state disputes reveals that the currently-dominant system for resolving investor-state disputes relies explicitly on only one method in one category: binding arbitration, in the imposed category. As noted earlier, a few states have begun to experiment with mechanisms that fit into the preventative category and that are available to an investor even before it begins to frame its concern as a "dispute,"75 or has to turn to arbitration. Now in the "next generation" phase of dispute system design, commentators agree that the best systems are characterized by the following:76 (1) multiple process options for parties, 77 including rights-based and interest-based processes; (2) ability for parties to "loop back" and "loop forward" among these options; (3) substantial stakeholder involvement in the system's design (with significant concern about the perceived unfairness of dispute system design systems designed by one disputing party and imposed upon the other

73. One of the confusing aspects of these terms in the investor-state context is that "conciliation" as used in the treaties sounds like a facilitated process but, as described in reality, appears to be closer to a nonbinding arbitration. See Salacuse, supra note 9, at 173. 74. Note that today's ADR procedures do not fit as neatly as they once did into the categories identified by Costantino and Merchant. See Lisa Blomgren Bingham, Collaborative Governance: Emerging Practices and the Incomplete Legal Framework for Public and Stakeholder Voice, 2009 J. Disp. RESOL. 269, 291 (2009) (explaining that processes for resolving conflict in policy-making vary along several dimensions, including the participants, their authority and power to influence policy decisions, and the process for communication and decision-making). 75. See, e.g., INVESTOR-STATE DISPUTES, supra note 7, at 49, 68 (discussing various initiatives taken by states to encourage prevention and negotiation of investor- state disputes, including Japan's state-state joint commissions; Malaysia's and the Philippines' broad consultations with their own governmental agencies during treaty negotiations; Peru's coordination and response system, including its Special Commis- sion; China's domestic administrative review process; and Colombia's lead agency approach). 76. See, e.g., Stephanie Smith & Jan Martinez, An Analytic Framework for Dis- pute Systems Design, 14 HARv. NEGOT. L. REV. 123, 128 (2009). 77. This element is consistent with the arguments of commentators who have promoted acceptance of "process pluralism." See, e.g., Carrie Menkel-Meadow, Peace and Justice:Notes on the Evolution and Purposesof Legal Processes, 94 GEo. L.J. 553, 554-56 (2006); John Lande, Getting the Faith: Why Business Lawyers and Executives Believe in Mediation, 5 HARv. NEGOT. L. REV. 137, 147-50 (2000). Harvard Negotiation Law Review [Vol. 18:71 disputing parties78 ); (4) participation that is voluntary, confidential, and assisted by impartial third party neutrals; (5) system transparency and accountability; 79 and (6) education and training of stakeholders on the use of available process options. Dispute system design scholarship originally posited that the initial focus in resolving disputes should be on interests, rather than rights or power.80 The current use of arbitration represents a movement from power (when some states bullied each other or bullied investors) to rights (since states and investors are treated as equal players, both bound by the terms of treaties and contracts). Media- tion, if understood as a presumptively interest-based technique, would represent the next movement, from rights to interests. But the more recent evolution of dispute system design no longer assumes that attempts at resolution must begin with an interest- based process. Instead, the best dispute systems simply include an interest-based process, and parties may begin with that process or another and loop forward and backward among the available processes. Meanwhile, as will be discussed infra, today's mediation process is no longer assumed to be exclusively interest-based; rights and power almost inevitably play a role. 8s So, integrating mediation into the investment treaty context would provide investors and states with the opportunity to resolve their disputes through a process that provides for explicit consideration of their interests, consistent with

78. See, e.g., Lisa Blomgren Bingham et al., Dispute System Design and Justice in Employment Dispute Resolution: Mediation at the Workplace, 14 HARv. NEGOT. L. REV. 1, 22-24 (2009). 79. See id. at 32-33 (describing the evaluation of the USPS REDRESS system based on users' perceptions of interactional justice (satisfaction with interpersonal treatment experienced during mediation), procedural justice (satisfaction with the process), and distributive justice (satisfaction with the outcome); Lisa Blomgren Bing- ham, Why Suppose? Let's Find Out: A Public Policy Research Program on Dispute Resolution, 2002 J. Disp. RESOL. 101, 115 (2002) (observing that USPS established percentage goals for voluntary employee participation which substantially influenced the design of the REDRESS program). 80. See URY ET AL., supra note 69, at 18. 81. See, e.g., Nancy A. Welsh, I Could Have Been A Contender: Summary Jury Trial as a Means to Overcome Iqbal's Negative Effects Upon Pre-Litigation Communi- cation, Negotiation and Early, Consensual Dispute Resolution, 114 PENN ST. L. REV. 1149, 1172-85(2010) [hereinafter Contender] (discussion of how rights and power influence the likelihood, tenor and results of communication, negotiation and assisted negotiation); Nancy A. Welsh, Stepping Back Through the Looking Glass: Real Con- versations with Real Disputants About Institutionalized Mediation and Its Value, 19 OHIO ST. J. ON Disp. RESOL. 573, 594-606 (2004) [hereinafter Stepping Back] (discussing how perceptions of power and rights influence expectations and perceptions of mediation); Salacuse, supra note 9. Spring 2013] Bilateral Investment Treaty Arbitration dispute system design principles, without eliminating consideration of rights. Throughout the years, dispute system design literature also has consistently emphasized stakeholders' role in designing the dispute system and the need to be able to demonstrate the system's positive impacts upon efficiency, effectiveness, stakeholders' satisfaction, and justice perceptions.8 2 Indeed, research suggests that stakeholders' engagement in decision-making regarding the design of a dispute system (including the processes that are included, the elements that are compulsory, and mechanisms to assure both informed stakeholder participation and system accountability), as well as their role in selecting the particular process or processes they will use to resolve their dispute and their subsequent experience with those processes, all impact their perceptions of the procedural (and substantive) justice offered by the system and individual processes. This Article will next turn, therefore, to a discussion of this research and theories of procedural justice.

B. ProceduralJustice Empirical research reveals that decision-making and dispute resolution procedures are most likely to be effective if they are perceived as procedurally fair.8 3 If parties perceive a dispute resolution or decision-making process as procedurally fair, they are more likely to perceive the outcome as substantively fair even if it is adverse to

82. It may be worthwhile to note that McAdoo and Welsh proposed a framework for evaluation that included efficient justice, substantive justice and procedural justice. See McAdoo & Welsh, supra note 63, at 425-29. See also Jeanne M. Brett, Ste- phen B. Goldberg, and William L. Ury, Designing Systems for Resolving Disputes in Organizations,45 American Psychologist 162, 169 (1990) (exploring the relevance of procedural justice to dispute system design, in terms of "incentives associated with process"). 83. See Nancy A. Welsh, Donna Stienstra & Bobbi McAdoo, The Application of Procedural Justice Research to Judicial Actions and Techniques in Settlement Ses- sions, in THE MULTI-TASKING JUDGE: COMPARATIVE JUDICIAL DISPUTE RESOLUTION (Tania Sourdin & Archie Zariski, eds., forthcoming 2013); Nancy A. Welsh, Percep- tions of Fairness in Negotiation, in THE NEGOTIATOR'S FIELDBOOK 165, 170 (Andrea K. Schneider & Christopher Honeyman, eds., 2006) [hereinafter Perceptions];Nancy A. Welsh, Making Deals in Court-ConnectedMediation: What's Justice Got To Do With It?, 79 WASH. U. L.Q. 787, 818 (2001) [hereinafter Making Deals]; see generally E. ALLAN LIND & TOM R. TYLER, THE SOCIAL PSYCHOLOGY OF PROCEDURAL JUSTICE 66-70, 205 (1988); E. Allan Lind, ProceduralJustice, Disputing,and Reactions to Legal Au- thorities, in EVERYDAY PRACTICES AND TROUBLE CASES 177, 192 (Austin Sarat, et al., eds., 1998); Tom R. Tyler, Social Justice: Outcome and Procedure, 35 INT'L J. PSYCHOL. 117, 119 (2000) [hereinafter Social Justice]; Tom R. Tyler & E. Allan Lind, ProceduralJustice, in HANDBOOK OF JUSTICE RESEARCH IN LAw, 65, 68 (Joseph Sand- ers & V. Lee Hamilton eds., 2001). HarvardNegotiation Law Review [Vol. 18:71 them,8 4 comply with that outcome,8 5 and perceive the institution that provides or sponsors the process as legitimate.8 6 Indeed, in the U.S., researchers have found that the public's overall approval of, and confidence in, the courts are influenced most strongly by their perception that the procedures offered by the courts are fair.8 7 Re- searchers have found that perceptions of procedural justice matter in decision-making processes as well as dispute resolution processes,88 in one-on-one negotiation8 9 as well as mediation 90 and arbitration, in

84. See LIND & TYLER, supra note 83, at 66-70, 205; Tyler, Social Justice: supra note 83, at 119. 85. See Tyler, Social Justice, supra note 83, at 119; Lind, supra note 83, at 192; Tom R. Tyler, Psychological Models of the Justice Motive: Antecedents of Distributive and ProceduralJustice, 67 J. PERSONALITY & Soc. PSYCHOL. 850, 857 (1994) [hereinafter Psychological Models]; Tom R. Tyler, Does the American Public Accept the Rule of Law? The Findings of PsychologicalResearch on Deference to Authority, 56 DEPAUL L. REV. 661, 660-70, 673-74 (2006-2007) [hereinafter Rule of Law] (describing procedural justice findings generally and research that has identified "procedural justice and trust as the key antecedents of the willingness to defer to legal authorities"). 86. See LIND & TYLER, supra note 83, at 209; ToM R. TYLER, WHY PEOPLE OBEY THE LAW 94-108 (1990); Lind, supra note 83, at 188; DAVID B. RoTrmAN, TRUST AND CONFIDENCE IN THE CALIFORNIA COURTS: A SURVEY OF THE PUBLIC AND ATTORNEYS 24 (2005); Tyler, Rule of Law, supra note 85, at 665. 87. Meanwhile, most lawyers focus on the outcomes of these procedures. See Tyler, Rule of Law, supra note 85, at 663 (reporting on research conducted in Califor- nia and elsewhere); RoiTMAN, supra note 86, at 25 (reporting that the public's evaluations of procedural fairness has greater influence on their evaluations of the courts, while attorneys give more weight to outcomes; also noting that studies in other states indicate that judges are also more concerned about outcome fairness than procedural fairness). 88. See LIND & TYLER, supra note 83, at 211-12, 242 (1988), cited in Welsh, Mak- ing Deals, supra note 83, at 817-18. 89. See Welsh, Perceptions, supra note 83, at 170; Welsh, Contender, supra note 81, at 1151-52; Rebecca Hollander-Blumoff & Tom R. Tyler, ProceduralJustice in Ne- gotiation: Procedural Fairness, Outcome Acceptance, and Integrative Potential, 33 LAw & Soc. INQUIRY 473, 477-79 (2008); Rebecca Hollander-Blumoff, Just Negotia- tion, 88 WASH. U. L. REV. 381, 385 (2010). 90. See, e.g., Roselle L. Wissler, Representation in Mediation: What We Know from Empirical Research, 27 FORDHAM URB. L.J. 419, 447-50 (2010) (exploring procedural justice effects of parties speaking in mediation compared to lawyers speaking); Roselle L. Wissler, Court-ConnectedMediation in General Civil Cases: What We Know from Empirical Research, 17 OHIO ST. J. ON DisP. RESOL. 641, 685 (2002) [hereinafter Court-Connected Mediation] (reporting procedural justice effects of lawyers' perceived cooperation in mediation); Tina Nabatchi et al., Organizational Justice and Work- place Mediation: A Six-Factor Model, 18 INT'L J. OF CONFLICT MGMT. 148, 164 (2007) (exploring procedural justice effects in transformative model of mediation); Nancy A. Welsh, Making Deals, supra note 83, at 838-58 (applying procedural justice research and theories to mediation); Nancy A. Welsh, Stepping Back, supra note 81, at 647-51, 661, 669-71 (suggesting that the use of caucus may magnify both positive and negative procedural justice perceptions). Spring 2013] Bilateral Investment Treaty Arbitration workplaces as well as courts, 91 and in countries with very different 9 2 cultures. Four process characteristics reliably predict parties' perceptions of fairness: the opportunity for parties to express themselves and their positions ("voice"), 93 demonstration of sincere consideration of 94 these expressions by a trustworthy decision-maker ("being heard"), even-handed treatment and the neutrality of the forum,95 and dignified, respectful treatment.9 6 Parties assess decision-makers'

91. See LIND & TYLER, supra note 83, at 211-12. But see Kristina A. Diekmann et al., Does FairnessMatter More to Some than to Others? The ModeratingRole of Work- place Status on the Relationship Between Procedural FairnessPerceptions and Job Satisfaction, 20 Soc. JUST. RES. 161, 163 (2007) (reporting research demonstrating that procedural justice has less influence with individuals of higher status in workplace); Jan-Willem Van Prooijen et al., Procedural Justice and Intragroup Status: Knowing Where We Stand in a Group Enhances Reactions to Procedures,41 J. EXPERI- MENTAL SOC. PSYCHOL. 644, 667-69 (2005); Jody Clay-Warner, PerceivingProcedural Injustice: The Effects of Group Membership and Status, 64 Soc. PSYCHOL. Q. 224, 227- 35 (2001). 92. See Welsh, Making Deals, supra note 83, at 821. 93. See LIND & TYLER, supra note 83, at 211-212; Lind, supra note 83, at 180; Tyler, Social Justice, supra note 83, at 121 (describing voice as the opportunity for people to present their "suggestions" or "arguments about what should be done to resolve a problem or conflict" or "sharing the discussion over the issues involved in their problem or conflict" and also noting that voice effects have been found even when people know they will have little or no influence on decision makers); Nourit Zimmerman & Tom R. Tyler, Between Access to Counsel and Access to Justice: A Psy- chological Perspective, 37 FoRDHAM URB. L.J. 473, 488 (2010) (reporting that voice "shapes evaluations about neutrality, trust, and respect" and has the "strongest influence, followed respectively by neutrality, trust, and respect"). 94. See Lind, supra note 83, at 179; Donald E. Conlon, et al., Nonlinear and Nonmonotonic Effects of Outcome on Procedural and Distributive Fairness Judg- ments, 19 J. APPLIED SOC. PSYCHOL. 1085, 1095 (1989); Tyler, Psychological Models, supra note 85, at 858; Tyler, Rule of Law, supra note 85, at 664; Tyler, Social Justice, supra note 83, at 122. 95. See Tyler, Rule of Law, supra note 85, at 664 (observing that "[t]ransparency and openness foster the belief that decision-making procedures are neutral"); see also Steven L. Blader & Tom R. Tyler, A Four-ComponentModel of Procedural Justice: Defining the Meaning of a "Fair"Process, 29 PERS. Soc. PSYCHOL. BUL. 747 (2003) (distinguishing between "formal" or "structural" aspects of groups that influence perceptions of process fairness, such as group rules, and the "informal" influences that result from individual authority's actual implementation of the rules); Welsh, Stien- stra & McAdoo, supra note 83 (suggesting "two quite different perspectives" regarding even-handed treatment, with the first perspective focusing on "interactional or relational matters" and the second perspective focusing on cognitive issues such as "whether or not the process provides a sufficiently informed and fact-based application of objective principles"). 96. See Tom R. Tyler, The Psychology of ProceduralJustice: A Test of the Group- Value Model, 57 J. PERSONALITY & Soc. PSYCHOL. 830, 831 (1989); E. Allan Lind, et al., In the Eye of the Beholder: Tort Litigants' Evaluations of Their Experiences in the Civil Justice System, 24 LAw & Soc'Y REV. 953, 948 (1990); Tyler, Social Justice, supra note 83, at 122. While respectful treatment is described here as an essential Harvard Negotiation Law Review [Vol. 18:71

trustworthiness 97 in order to determine whether they "can trust that in the long run the [decision-making] authority with whom they are dealing will work to serve their interests."98 Perhaps because parties

element of procedural justice, it has also been described as an element of interactional justice, and even of distributive justice. See Robert J. Bies, Are ProceduralJustice and Interactional Justice Conceptually Distinct?, in HANDBOOK OF ORGANIZATIONAL JUSTICE 85 (Jerald A. Greenberg & Jason A. Colquitt, eds., 2005). See also Welsh, Perceptions, supra note 83, at 170; Nancy A. Welsh, Remembering the Role of Justice in Resolution: Insights from Procedural and Social Justice Theories, 54 J. LEGAL EDUC. 49, 52 (2004); Welsh, Making Deals, supra note 83, at 820-21. Tom R. Tyler, one of the most prominent procedural justice researchers, has described these four elements slightly differently: What makes a process fair in the eyes of the public? Four critical factors dominate evaluations of procedural justice. First, people want to have an opportunity to state their case to legal authorities. They want to have a forum in which they can tell their story; they want to have a "voice" in the decision-making process. Second, people react to signs that the authorities with whom they are dealing are neutral. Neutrality involves making decisions based upon consistently applied legal principles and the facts of the case rather than personal opinions and biases. Transparency and openness foster the belief that decision-making procedures are neutral. Third, people are sensitive to whether they are treated with dignity and politeness and to whether their rights as citizens and as people are respected. Finally, people focus on cues that communicate information about the intentions and charac- ter of the legal authorities with whom they are dealing. People react favora- bly to the perception that the authorities are benevolent and caring and are sincerely trying to do what is best for the public-that is, when they trust that authority. Authorities communicate this concern when they listen to people's accounts and explain or justify their actions in ways that show an awareness of people's needs. Tyler, Rule of Law, supra note 85, at 664. 97. Lind, supra note 83, at 192; see also Tyler & Lind, supra note 83, at 768 (re- framing "trust" as "trust in benevolence" and defining this relational term as "infer- ences about the authority's motivations, especially the authority's willingness to consider one's needs and to try to make fair decisions"). 98. Tyler, PsychologicalModels, supra note 85, at 854. A key question, of course, is what serves as the ultimate basis for people's assessment of whether a decision- maker is well-meaning and trustworthy. One of the authors of this article has suggested a meaningful correlation among the behaviors that signal procedural justice, a cooperative style, and perceived trustworthiness. If a decision-maker behaves in a procedurally just manner and has a cooperative style, this tends to be correlated with trustworthiness, and the resulting atmosphere of trust has been shown to generate greater willingness to share the sort of information that leads to mutually-beneficial, integrative solutions. See Nancy A. Welsh, ReputationalAdvantages, supra note 63, at 132-33 (citing to Hollander-Blumoff & Tyler, supra note 89. This line of research suggests that the outcomes produced by a decision-maker who is perceived as well- meaning and trustworthy are likely to be viewed as fair and also likely to be more integrative. But this effect has also been shown to be moderated by pre-existing perceptions of the decision-maker's trustworthiness-or lack of trustworthiness. In other words, procedurally just behaviors will not have their usual effects if the decision-maker is distrusted; they will have their usual effects when there is no information about the trustworthiness of the decision-maker and their effects may be enhanced when the decision-maker is trusted. See David De Cremer & Tom R. Tyler, Spring 2013] BilateralInvestment Treaty Arbitration realize that these procedural characteristics can be manipulated, however, they tend to be on high alert for "sham" procedures. 99 For fairly obvious reasons, parties are likely to be particularly vigilant regarding the potential for a "sham" when they are uncertain that they can trust the others involved a dispute resolution process and/or the dispute is a very serious one, involving the potential for grievous harm.100 Several theories explain why parties care so much about procedural justice. First, parties want to be reassured that the decision- maker has access to, and considers, the information they present. If the decision-maker has this information, and demonstrates consideration of it, parties are more willing to believe that their interests will be protected. 10 1 Indeed, because it can be so difficult to determine whether an outcome is substantively fair, some have theorized that parties' judgment regarding the fairness of a procedure acts as a heuristic for their judgments regarding the fairness of outcomes.102 Sec- ond, the procedures themselves communicate whether the parties

The Effects of Trust in Authority and Procedural Fairness in Cooperation, 92 J. OF APPLIED PSYCHOL. 639, 646-47 (2007); see also David Markell, et al., What's Love Got To Do With It?: Sentimental Attachments and Legal Decision-Making,57 VILL. L. REV. 209, 239-40 (2012) (reporting research finding that trust in the motives of authorities is primary when sentimental values are at stake, while perceiving a decision-maker to be neutral is most important when monetary interests are primary). 99. See Lind, supra note 83, at 187 (observing that if people perceive evidence of unfair treatment or perceive "false representations of fair treatment," they respond with "extremely negative reactions"); Tom R. Tyler, et al., Influence of Voice on Satis- faction with Leaders: Exploring the Meaning of Process Control, 48 J. PERSONALIY & SOC. PSYCHOL. 72, 74 (1985) (explaining that, under certain conditions, voice without decision control heightens feelings of procedural injustice and dissatisfaction with leaders, a result described as the "frustration effect"). 100. See Welsh, Reputational Advantages, supra note 63, at 136-138 (regarding the relationship between perceptions of procedural justice and trustworthiness, and the appropriate incidence of distrust); Andrea K. Schneider, Perception, Reputation and Reality: An Empirical Study of Negotiation Skills, 6 Disp. RESOL. MAG. 24 Sum- mer 2000, at 24. 101. This is described as the "social exchange" theory. People wish to know that the decision maker is fully informed regarding their perspective, in hopes that this will influence the outcome. See Lind, supra note 83, at 179. Recent research also interestingly suggests that in negotiation, the creation of integrative solutions is correlated with procedurally just behaviors. See Hollander-Blumoff & Tyler, supra note 89, at 477-79. 102. See E. Allan Lind et al., Individual and CorporateDispute Resolution: Using Procedural Fairness as a Decision Heuristic, 38 ADMIN. Sci. Q. 224, 225 (1993) (reporting that researchers found that procedural justice judgments strongly influenced litigants' decisions about whether or not to accept nonbinding arbitration awards, regardless of whether litigants were individuals, small business owners, or corporate officers; only corporate employees demonstrated no link between their procedural justice judgments and their decisions to accept awards). Harvard Negotiation Law Review [Vol. 18:71 accessing those procedures are deserving of respect. If the neutral in a dispute resolution process listens to the parties before her and consistently demonstrates both respect and a sincere attempt to be open- minded and even-handed, these behaviors signal to the parties that they are valued members of the group, regardless of whether that group is a nation, a local community, or a workplace.103 Refusal to listen or closed-mindedness signals a lack of respect. More recently, Allan Lind and others have urged that parties use their perceptions of procedure as a mechanism to manage the negative dynamics, sense of vulnerability, and risk often associated with uncertainty. 10 4 A fair procedure communicates the decision maker's (and the sponsoring institution's) respect for, and well-meaning attitude toward, the party which can then help to reduce the anxieties associated with actual loss, feared loss, and/or an uncertain future. Recent research has also revealed that although procedural justice matters to most people, it can matter to some people more than others. For example, those who perceive themselves as having lower or uncertain status are more likely to perceive a just outcome if the higher status decision-maker-who could be the neutral or the other negotiator-treats them in a procedurally just manner.10 5 Parties who are collectivists or who find themselves in situations that accen- tuate hierarchy and unequal status106 are also likely to be very aware if they are treated in a procedurally just manner.107 Individu- alists and higher status parties, in contrast, are much less influenced by procedural justice. Indeed, their positive perceptions regarding a

103. This is the "group value" or "relational" theory. People notice the psychological message that procedures convey regarding their value to the relevant social group. To receive respect and sincere consideration signals the individual's value and social standing. See Tyler, Psychological Models, supra note 85, at 858. 104. See Hollander-Blumoff & Tyler, supra note 89, at 477 (citing E. Allan Lind, Fairness Judgments as Cognitions, in THE JUSTICE MOTIVE IN EVERYDAY LIFE, (Michael Ross & Dale T. Miller eds., 2002)); Kees van den Bos & E. Allan Lind, Uncer- tainty Management by Means of FairnessJudgments, 34 ADVANCES IN EXPERIMENTAL SOC. PSYCHOL. 1, 26-30 (2002). See also Nancy A. Welsh & Barbara Gray, Searching for a Sense of Control: The Challenge PresentedBy Community Conflicts Over Concen- trated Animal Feeding Operations, 10 PENN ST. ENvTL. L. REV. 295 (2002). 105. See Welsh, Perceptions, supra note 83, at 170-71. 106. See id. at 166. 107. See id. at 170 (citing to Jan Wilhelm Van Prooijen et al., ProceduralJustice and Status Salience as Antecedent of ProceduralFairness Effect, 83 J. PERSONALITY & SOC. PSYCHOL. 1353, 1359 (2002) ); see also Clay-Warner, supra note 91, at 227-35; Diekmann et al., supra note 91, at 163; Van Prooijen et al., supra note 91, at 667-69. Spring 2013] Bilateral Investment Treaty Arbitration process will matter less than the "bottom line"-i.e., whether the outcome is at least consistent with their expectations.10 8 Procedural justice research is particularly important in the investment treaty context, as some states threaten noncompliance and as all stakeholders express a desire to know that they are being treated fairly within a system that they perceive as legitimate. It is obviously important that states and investors perceive the investment treaty arbitrationprocess as procedurally just.10 9 Procedural justice theories and research can provide useful procedural benchmarks to arbitrators and arbitral organizations committed to

108. See JANE ADLER ET AL., SIMPLE JUSTICE: How LITIGANTS FARE IN THE PITTS- BURGH COURT ARBITRATION PROGRAM, 61-62 (1983) (discussing difference between organizational and individual parties' reactions to Pittsburgh arbitration program); Lind et al., supra note 102, at 247 (reporting that procedural justice judgments strongly influenced litigants' decisions whether or not to accept non-binding arbitration awards, regardless of whether litigants were individuals, small business owners, or corporate officers, except that corporate employees failed to demonstrate such link); Wayne Brazil, Hosting Mediations as a Representative of the System of Civil Justice, 22 OHIO ST. J. ON Disp. RESOL. 227, 237-38 (2007) (expressing no surprise that "big- time economic actors" would acquire thicker "'process skin'" and be "much more concerned about ends than means ... [and thus] not likely to mind a little 'process rough- ness' if they sense that it increases the odds that they will get a deal"); Tyler, Social Justice, supra note 83, at 123 (describing the significance of social categorization and referencing research showing that "people are less concerned about justice when they are dealing with people who are outside their own ethnic or social group;" and "when people have a dispute with someone who is not a member of their own social group, they pay more attention to the personal favourability of a proposed dispute resolution when deciding whether to accept it"); Diane Sivasubramaniam & Larry Heuer, Deci- sion Makers and Decision Recipients: UnderstandingDisparities in the Meaning of Fairness,44 CT. REV. 62, 66 (2007-2008) (reporting several experiments that demonstrated that those assuming the role of authority or decisionmaker were more likely to define fairness in terms of outcome, while those who were decision recipients were more likely to be concerned with respectful, fair treatment). But see Donna Shestow- sky & Jeanne Brett, Disputants' Perceptions of Dispute Resolution Procedures:An Ex Ante and Ex Post Longitudinal Empirical Study, 41 Con. L. REV. 63, 94-106 (2008) (finding that those who expressed pre-process preference for a process in which a third party made the decision were likely to be satisfied with that process, and detailing research indicating that corporations prefer mediation due to their ability to control outcome, which can be understood as being consistent with achieving expectations). 109. See Susan D. Franck, The ICSID Effect? Considering PotentialVariations in ArbitrationAwards, 51 VA. J. INT'L L. 825 (2011) (testing the legitimacy of ICSID by using empirical analysis to demonstrate that ICSID results are not different than other arbitral processes); Dohyun Kim, The Annulment Committee's Role in Multiply- ing Inconsistency in ICSID Arbitration: The Need to Move Away From an Annulment- Based System, 86 N.Y.U. L. REV. 242, 255-56 (2011) (outlining social legitimacy as decisions that are perceived as justified and that states accept are justified); William W. Burke-White, The Argentine FinancialCrisis: State Liability Under BITs and the Legitimacy of the ICSID System, 3 ASIAN J. WTO & INT'L HEALTH L. & POL'Y 199, 288 (2008). Harvard Negotiation Law Review [Vol. 18:71 achieving these goals. 110 Perceptions regarding the procedural justice of investment treaty mediation, however, will also matter.111 In fact, perceived and actual procedural justice should be the goal for all of the dispute resolution procedures that comprise the dispute resolu- 112 tion system available in the investment treaty context. Further, we should take a step back to examine the decision- making process that leads to the development of the dispute resolution clauses in investment treaties, including such clauses' definition of the array of available processes and the mechanism that will determine the process to be used for a particular dispute. Research suggests that stakeholders' perceptions of procedural justice are likely to matter just as much in this "upstream" decision-making context as in the later "downstream" dispute resolution process." 3 Professor Lisa

110. See, e.g., Roger K. Warren, Public Trust and ProceduralJustice, 37 CT. REV. 3, 12, 15 (2000) (describing procedural justice-related performance criteria used in court-sponsored evaluation programs and proposing additional measurement of trustworthiness); Judge Kevin Burke & Judge Steve Leben, ProceduralFairness: A Key Ingredient in Public Satisfaction:A White Paper,44 CT. REV. 1/2, 4-5 (2007). The Na- tional Center for State Courts offers CourTools, Trial Court Performance Measures that include a series of questions focusing on litigants' perceptions regarding access to the courts and the fairness of their procedures. See CourTools: Giving Courts the Tools to Measure Success (2005) [hereinafter CourTools], available at http:ll www.courtools.org/-/media/Microsites/Files/CourTools/ CourToolsTrialBrochure.ashx 111. See Welsh, Making Deals, supra note 83, at 170; Welsh, Stepping Back, supra note 81, at 606; McAdoo & Welsh, supra note 63, at 410, 416, 420, 423-27 (discussing data that supports and does not support court-connected mediation's achievement of procedural justice and making recommendations based on such data); Bobbi McAdoo, A Mediation Tune-Up for the State Court Appellate Machine, 2010 J. Disp. RESOL. 327, 328-32 (2010) (describing procedural justice and suggesting that appellate mediation can be structured to achieve both settlement and procedural justice). 112. See Welsh, Making Deals, supra note 83, at 170; see also Nancy A. Welsh, The Place of Court-Connected Mediation in a Democratic Justice System, 5 CARDozo J. CONFLICT RESOL. 117, 124-33 (2004) [hereinafter Place of Mediation]; Nancy A. Welsh, What is "(I'm) PartialEnough" in a World of Embedded Neutrals?, 52 ARIz. L. REV. 395, 423-28 (2010) [hereinafter Embedded Neutrals]. 113. See Bingham, supra note 68, at 74. See also Lisa Blomgren Bingham, Design- ing Justice: Legal Institutions and Other Systems for Managing Conflict, 24 OHIO ST. J. ON Disp. RESOL. 1, 5 (2008) (using Elinor Ostrom's seven categories for institutional design analysis and outlining a table/tautology of different types of justice); Keith G. Allred, RelationshipDynamics in Disputes:Replacing Contention with Cooperation, in THE HANDBOOK OF DISPUTE RESOLUTION, supra note 65, at 83, 91 (Michael L. Moffitt & Robert C. Bordone eds., 2005) [hereinafter HANDBOOK OF DISPUTE RESOLUTION] ("If we are in a dispute with someone and he or she asks us what we think the process should be to try to resolve it, we will feel the process is fairer whatever the process turns out to be"); Leonard L. Riskin & Nancy A. Welsh, Is That All There Is?: The "Problem"in Court-Oriented Mediation, 15 GEo. MASON L. REV. 863, 928 (2008) (proposing that courts should allow parties to customize court-connected mediation and Spring 20131 Bilateral Investment Treaty Arbitration

Bingham has noted, "[iln its best practice, DSD... uses inclusive, participatory, stakeholder-driven processes to change existing or create new dispute resolution structures. Its goal is to improve the capacity of systems to prevent, manage, or resolve certain streams or kinds of conflict." 114 Stakeholders are likely to perceive procedural justice in this sort of "inclusive, participatory" process, used to design or amend the dispute resolution clause in an investment treaty, if and only if they receive the opportunity for voice, serious and trustworthy consideration, and even-handed, dignified treatment in a neutral forum. 11 5 In other words, their perceptions of procedural justice will depend upon how their participation is managed. Such perceptions will matter because they will influence stakeholders' perceptions regarding the substantive justice of the treaty's dispute resolution clause and prescription of particular procedures. It will also impact the likelihood of the stakeholders' compliance with the treaty provisions and their respect for the legitimacy of the states engaged in making the treaty. Thus, attention to procedural justice should enhance the effectiveness of the participatory stakeholder processes prescribed by dispute system design. 116 urging that such opportunity will enhance parties' perceptions of procedural fairness of process). 114. Bingham, supra note 68, at 75; see also URY ET AL., supra note 69, at 65-83 (1988); John Lande, Using Dispute System Design Methods to Promote Good-Faith Participation in Court-Connected Mediation Programs, 50 UCLA L. REV. 69, 115 (2002). 115. See also Nancy A. Welsh & Barbara Gray, Searching for a Sense of Control: The Challenge PresentedBy Community Conflicts Over Concentrated Animal Feeding Operations, 10 PENN ST. ENVTL. L. REV. 295,303-5 (2002). Note that "participation" is sometimes used as a synonym for "voice." See, e.g., Zimmerman & Tyler, supra note 93, at, 487 ("a study of small claims courts found that the opportunity for direct, un- mitigated participation-especially the opportunity for the litigant to tell his story before a judge-was an important determinant of the litigant's satisfaction"). But active "participation" in a decision-making process is likely to require something more than just "voice." It requires give-and-take, and listening as well as expressing one's own point of view. See Welsh, Stepping Back, supra note 81, at 606. Researchers have found that while mediating parties' perceptions of procedural justice are enhanced by the opportunity to "tell their views," these perceptions are not affected by the opportunity to "participate" in the process. This has led Roselle Wissler to suggest that "parties' sense of voice is more important to their experience in mediation than is how much they participate." Wissler, supra note 90, at 450. 116. Of course, the devil is in the details. The stakeholders invited to participate in a DSD process should include not just the official representatives of the states responsible for negotiating an investment treaty, but also representatives of the local and regional regulatory bodies that will be obligated to uphold (or avoid violating) the states' treaty obligations. Similarly, the process should include respected representatives of both large (and arguably multinational) foreign investors and the much smaller investors who may choose to invoke the new system. See Salacuse, supra note Harvard Negotiation Law Review [Vol. 18:71

Much later, when a particular dispute emerges, dispute system design's preference for loop-backs and loop-forwards suggests that the designated dispute resolution facility should provide the disputing state actors and investors with another opportunity for input- into the selection of the particular process that will be used to resolve their dispute (including, if appropriate, the particular model of that process), 117 the timing of such process, and the particular neutral or neutrals who will conduct the process. Again, the opportunity for such input 118 is likely to have positive effects in terms of procedural justice perceptions, provided that the parties believe that their input is being received respectfully, given serious and trustworthy consideration, and judged in an even-handed manner in a neutral forum." 9 With this brief introduction to dispute system design and procedural justice, this Article will now turn to an examination of the experience with court-connected and court-oriented mediation in the U.S. This examination will reveal significant variations among mediation models. Only some of these variations are different enough from other available procedures (especially conciliation) to meet dispute

9; see also Chris Carlson, Convening, in THE CONSENSUS BUILDING HANDBOOK: A COM- PREHENSIVE GUIDE To REACHING AGREEMENT 169 (Lawrence Susskind et al., eds., 1999) (discussing the convening function); BARBARA GRAY, COLLABORATING: FINDING COMMON GROUND FOR MULTIPARTY PROBLEMS 261-7 (1989); BERNARD MAYER, THE DY- NAMICS OF CONFLICT RESOLUTION: A PRACTITIONER'S GUIDE 225 (2000); Laurel S. Terry, From GATS to APEC: The Impact of Trade Agreements on Legal Services, 43 AKRON L. REV. 875, 888-89 (2010) (demonstrating the value of "conversation starter" provisions in international trade agreement that require the development of cross- border professional services working groups and have resulted in the active participation of state judiciaries responsible for the regulation of lawyers). 117. See Riskin & Welsh, supra note 113, at 919-21 (proposing that U.S. courts and private dispute resolution providers should provide parties with the opportunity to customize their mediation process and describing some courts' current processes for such customization); McAdoo & Welsh, supra note 44 (explaining effectiveness of some courts' voluntary mediation programs and emphasizing the role of court person- nel who reach out to parties). 118. The state will almost inevitably have received its opportunity for direct input when the treaty was being drafted and the mandated dispute system was being designed. Even if investor representatives and selected local government officials also were engaged in the treaty-drafting process, the individual investor and local officials involved in a particular dispute might not receive the opportunity for direct input until much later. See Bingham, supra note 78, at 22-24 (discussing the importance of determining who will choose the particular process to be used to resolve a dispute); see also Riskin & Welsh, supra note 113, at 869 (discussing issues on customization). But see Chris Guthrie, Panacea or Pandora'sBox, 88 IowA L. REV. 601 (discussing the perils of having too many choices); Jean Sternlight, In Defense of Mandatory Arbitra- tion (If Imposed on the Company), 8 NEv. L.J. 82 (2007). 119. But recall, once again, that the disputing parties are likely to be quite vigilant for signs that their participation is actually just a "sham." See Welsh, Embedded Neutrals, supra note 112, at 423-28. Spring 2013] Bilateral Investment Treaty Arbitration system design's prescription for multiple process options, interest- based processes as well as processes based on rights and power, and the need for meaningful loop-backs and loop-forwards. The Article will also examine the many variations among compulsory mediation referral schemes in order to find those few that are most likely to meet dispute system design's prescription for stakeholder involvement as well as the opportunity for voice, serious and trustworthy consideration, and even-handed, neutral and dignified treatment that lead to procedural justice perceptions. Finally, the Article will discuss potential quality controls in the selection and performance of the pool of mediators, to provide for accountability pursuant to dispute system design.

IV. THE MEDIATION PROCESS, REFERRAL SCHEMES AND QUALITY CONTROL MECHANISMS A. The Mediation Process 1. Basic Differences between Mediation and Arbitration It is a truism that mediation and arbitration are different. But closer examination reveals that mediation is substantively different from arbitration-i.e., binding arbitration12 0-in only one key respect: the neutral's degree of control over the outcome. 121 In mediation, the neutral (or mediator) assists the parties with their communications and negotiation. She cannot impose a solution. If there is a binding resolution reached in mediation, it will be the result of the parties' voluntary agreement.1 22 In contrast, at least in binding arbitration, decision-making power vests in arbitrators alone. They have the authority to decide outcomes for parties and

120. On occasion, researchers have treated mediation and non-binding arbitration as essentially the same. See, e.g., Tom R. Tyler, Citizen Discontent with Legal Proce- dures:A Social Science Perspectiveon Civil ProcedureReform, 45 AM. J. COMP. L. 871, 876-78 (1997). 121. Researchers have found that this distinction matters. See e.g., Shestowsky & Brett, supra note 108, at 94-106. (reporting research showing that disputant choice between mediation and arbitration is based largely on desired level of third party control over the outcome) 122. Of course, case law suggests that this ideal is not always met. See Nancy A. Welsh, The Thinning Vision of Court-Annexed Mediation: The Inevitable Price of In- stitutionalization?,6 HARv. NEGOT. L. REV. 1, 17-19 (2001) [hereinafter Thinning Vi- sion]; see also Nancy A. Welsh, Disputants' Decision Control in Court-Connected Mediation: A Hollow Promise Without Procedural Justice, 2002 J. Disp. RESOL. 179, 181-84(2002) [hereinafter Hollow Promise]; James R. Coben & Peter N. Thompson, DisputingIrony A Systematic Look at LitigationAbout Mediation, 11 HARv. NEGOT. L. REV. 43, 127-38 (2006). HarvardNegotiation Law Review [Vol. 18:71 issue binding awards. 123 Indeed, in the U.S. domestic context, arbitral awards often appear more binding than those issued by many courts. And, in the international context, there is no question that arbitration awards are easier to enforce than court judgments or ne- 124 gotiated agreements. In common law countries, arbitration also tends to have a different "look" than mediation. While mediation looks more like an informal meeting, binding arbitration mimics a more formal judicial hearing. 125 The arbitration process can feature opening statements, direct and cross examination of witnesses, determinations regarding the admissibility of documents into evidence, and even closing arguments. While the arbitrators may interrupt the parties' presentations to ask questions, the parties present their cases in sequential order. If lawyers are involved, the parties and witnesses tend to tes- tify only in response to the lawyers' or the arbitrators' questions. Each party's case is presented in the presence of the other party. The process is structured to ensure that there is transparency, at least as between the parties,1 26 and that the arbitrators have the information needed to make their decision, which will be binding upon the parties.127 In contrast to the comparatively formal arbitration process, mediation in these common law countries is a much less-obviously structured affair. The process often, but not always, begins with pre- mediation submissions and telephonic conferences with the mediator. On the day of the mediation, the parties meet in a conference room,

123. Note, however, that there are also important variations on binding arbitration that make the size of the award or its finality contingent party choice. For example, in non-binding arbitration a party may choose to reject the award and proceed to a different adjudicative process or litigation. In high-low arbitration or baseball arbitration, the parties agree to constrain the arbitrator's discretion in making a monetary award. 124. But see Edna Sussman, The New York Convention Through a Mediation Prism, 15 Disp. RESOL. MAG., 10 Summer 2009 (referencing other nations' approaches for enforcing judicially-brokered agreements). 125. But see Elsing, supra note 59, at 313 (describing fast-track arbitration, chess clock arbitration, and guillotine arbitration). 126. See Coe, Transparency, supra note 7, at 1345 n. 32 (describing various measures undertaken to make the NAFTA arbitration regime more transparent for non- parties-e.g., presumptively open arbitration hearings, participation by non-disputant states and amici). 127. See LEONARD L. RISKIN, ET AL., DISPUTE RESOLUTION AND LAwYERS 557-58 (4th ed. 2009) (discussing the role of the arbitrator, and the components needed for a successful arbitration); CARRIE MENKEL-MEADOW ET AL, DISPUTE RESOLUTION: BE- YoND THE ADVERSARIAL MODEL 483-491(2nd ed, 2011) (discussing arbitration skills and typical format). Spring 20131 Bilateral Investment Treaty Arbitration and the process often (but not always) begins with an orientation by the mediator and opening statements by each side. The mediators and parties may remain together throughout the process but are more likely to separate into different rooms, with the mediator "shuttling" among them. If the parties have separated and reach resolution, they may reconvene for a joint meeting to confirm the terms of the settlement. The process is structured to permit the parties to engage in joint and direct communication with each other, but also private deliberation and, through the medium of ex parte meetings with the mediator, indirect communication with each other.128 Thus structured, the process ensures that the parties-rather than a third party decision-maker-have sufficient information to allow them to reach a decision that they can accept and will implement. Last, mediation tends to differ from arbitration in its explicit consideration of the parties' interests. Indeed, some courts in the U.S. describe mediation as follows: Mediation is a process in which parties and counsel agree to meet with a neutral mediator trained to assist them in settling disputes. The mediator improves communication across party lines, helps parties articulate their interests and understand those of the other party, probes the strengths and weaknesses of each party's legal positions, and identifies areas of agreement and helps generate options for a mutually agreeable resolution to the dispute. In all cases, mediation provides an opportunity to explore a wide range of potential solutions and to address interests that may be outside the scope of the stated controversy or which could not be addressed by judicial action. A hallmark of mediation is its capacity to expand traditional settlement dis- cussions and broaden resolution options, often by exploring litigant needs and interests that may be formally independent of 129 the legal issues in controversy.

128. See Jennifer G. Brown & Ian Ayres, Economic Rationales for Mediation, 80 VA. L. REV. 323, 325-26 (1994); Ian Ayres & Barry J. Nalebuff, Common Knowledge as a Barrier to Negotiation, 44 UCLA L. REV. 1631, 1633-34 (1997); Welsh, Stepping Back, supra note 81, at 647-51, 660-71; Nancy A. Welsh, Musings on Mediation, Kleenex, and (Smudged) White Hats, 33 U. LAVERNE L. REV. 5, 13 n. 42 (2011) [hereinafter Musings] (observing that mediators' ability to engage in ex parte communications with parties is emerging as one of the process's "most-prized attributes"). 129. S.D.N.Y. & E.D.N.Y. Civ. R. 83.8 (2011) (applying exclusively to the Eastern District). Harvard Negotiation Law Review [Vol. 18:71

Thus, in general, mediation offers the opportunity for parties to discuss both their legal and extra-legal issues and needs. In the investor-state context, these may include financial constraints or aspi- rations, domestic political realities, regional concerns, and protection of important community norms or characteristics.

2. A More Nuanced Appreciation of the Relationships Among Models of Mediation, Arbitration, and Other Processes

It is easy and tempting, however, to overstate the differences between mediation and arbitration. In fact, the differences may not be obvious at all in terms of outcome, "look," or focus. Regarding the neutral's degree of control over the outcome, for example, the lawyers and parties in a nation with a common law tradition like the U.S. may be willing to grant substantial deference to a mediator's reactions and opinions, thus giving the mediator near-control over the outcome. This is likely to be especially true if the lawyers and parties have selected the mediator for his substantive knowledge, relevant mediation experience, status, facility for managing the process, perceived even-handedness, and general temperament. In other words, even in common law countries, a mediator may represent a "respected elder" within the relevant legal and business communities.130 Though the mediator does not have the authority to impose a solution, his assessments and suggestions have the potential to be extremely influential. 131 For the parties, meanwhile, sharing outcome control with an adversary may not feel substantially different 32 from ceding control to a third party decisionmaker.1

130. Christopher Honeyman & Sandra Cheldelin, Have Gavel, Will Travel: Dis- pute Resolution's Innocents Abroad, 19 CONFLICT RESOL. Q. 363, 365-7 (2002); see also Salacuse, supra note 9, at 35. 131. See Stephen B. Goldberg & Margaret L. Shaw, The Secrets of Successful (and Unsuccessful) Mediators, 8 Disp. RESOL. ALERT 1, 5-6 (2008) [hereinafter Goldberg & Shaw I]; Stephen B. Goldberg & Margaret L. Shaw, The Secrets of Successful (and Unsuccessful) Mediators Continued: Studies Two and Three, 23 NEGOT. J. 393, 407 (2007) [hereinafter Goldberg & Shaw II]; Riskin & Welsh, supra note 113, at 874 (citing research studies showing that lawyers prefer mediators who are litigators with relevant substantive expertise who will provide their opinions regarding the merits of the parties' cases as well as a suggested settlement range and that cases are significantly more likely to settle in mediation if the parties select their own mediator). 132. See Welsh, Hollow Promise, supra note 122, at 183. Spring 2013] BilateralInvestment Treaty Arbitration 109

Similarly, the strength of the distinction between arbitration and mediation in terms of their "look" is likely to depend upon the particular models of arbitration and mediation that are used. 133 In some models of mediation, for example, the mediator focuses on encouraging the disputing parties to express themselves, understanding their values and underlying interests, helping them to communicate fully, respectfully, and productively with each other, asking them questions to help them be realistic, and fostering their ability to develop their own customized solutions.134 These models tend to be called "facilita- 3 8 tive,"1 35 "elicitive,"' 36 "understanding-focused,"' 37 "therapeutic,"'

133. Other scholars have used different taxonomies to distinguish among different mediator behaviors. See, e.g., Salacuse, supra note 9, at 160 (distinguishing the following "basic areas that mediators seek to address in their efforts to facilitate a negotiated agreement between the parties: a) process, b) communications, and c) substance"); Jacob Bercovitch & Scott Sigmund Gartner, Is There Method in the Mad- ness of Mediation?: Some Lessons for Mediators from Quantitative Studies of Media- tion, in INTERNATIONAL CONFLICT MEDIATION: NEW APPROACHES AND FINDINGS 19, 27, 36-37 (Jacob Bercovitch & Scott Sigmund Gartner eds., 2009) (identifying "three fundamental mediator strategies along a continuum ranging from low to high interven- tion . . . (a) communication-facilitation; (b) procedural strategies; and (c) directive strategies" and reporting that international mediators, with significant resources and high status and using directive strategies, are most likely to achieve settlements in high-intensity disputes, while regional mediators, with cultural similarity and using procedural strategies, are most likely to achieve settlements in low-intensity conflicts); Goldberg & Shaw I, supra note 131, at 3; Goldberg & Shaw II, supra note 131, at 394-98, 407-08, 410-13; Nadja Alexander, The Mediation Meta Model: Understand- ing Practice,26 CONFLICT RESOL. Q. 97, 98 (2008) (proposing the following categories of mediation: settlement mediation, facilitative mediation, transformative mediation, expert advisory mediation, wise counsel mediation, and tradition-based mediation); John Wood, Mediation Styles: Subjective Description of Mediators, 21 CONFLICT RESOL. Q. 437, 447-8 (2004) (describing various styles as counselor, negotiator, facilitator and democrat). 134. Riskin & Welsh, supra note 113, at 869; Welsh, Thinning Vision, supra note 122, at 17-20. 135. See Leonard L. Riskin, Understanding Mediators' Orientations, Strategies, and Techniques: A Grid for the Perplexed, 1 HARv. NEGOT. L. REV. 7, 24 (1996) (discussing mediation in facilitative form where the mediator clarifies and enhances communications between parties in order to help them come to decision). 136. See Leonard L. Riskin, Decisionmaking in Mediation: The New Old Grid and the New New Grid System, 79 NOTRE DAME L. REV. 1, 23 (2003) (explaining that mediation can be performed in elicitive fashion). 137. See GARY FRIEDMAN & JACK HIMMELSTEIN, CHALLENGING CONFLICT: MEDIA- TION THROUGH UNDERSTANDING (2008). 138. See Susan S. Silbey & Sally E. Merry, Mediator Settlement Strategies, 8 LAw & POL'y 7, 12, 19 (1986) (suggesting mediation process is similar to therapeutic event); see also CHRISTOPHER W. MOORE, THE MEDIATION PROCESS: PRACTICAL STRATE- GIES FOR RESOLVING CONFLICT, 41 (2d ed. 1996) (describing other categories such as "social network mediators," "authoritative mediators," and "independent mediators'"-categories that have more to do with relationships between mediator and disputants than particular types of interventions that they tend to use). 110 Harvard Negotiation Law Review [Vol. 18:71

"humanistic,"139 "narrative,"'140 "insight,"' 41 and "transformative."1 42 Such models permit the parties to play a central role if they wish, though they also may choose to defer to their lawyers. There are meaningful differences among these models, but they tend to value communication among the parties, reflective listening, and the use of 143 joint sessions for as long as they are productive. Particularly in complex matters, whether they are commercial, environmental, or court-connected, mediators using these models are likely to engage in substantial "pre-mediation" work. These mediators are likely to review court documents, require pre-mediation submissions from the parties, and confer with the lawyers (and

139. See Mark S. Umbreit, Humanistic Mediation: A Transformative Journey of Peacemaking (1997) available at http://www.cehd.umn.edu/ssw/rjp/resources/rj_dia- logue-resources/humanistic approach/HumanisticMediationTransormativeJour- ney.pdf. 140. See JOHN WINSLADE & GERALD MONK, NARRATIVE MEDIATION: A NEW AP- PROACH TO CONFLICT RESOLUTION (2000). 141. See Cheryl Picard & Marnie Jull, Learning Through Deepening Conversa- tions: A Key Strategy of Insight Mediation, 29 CONFLICT REsOL. Q. 151, 152 (2011). 142. See ROBERT A. BARUCH BUSH & JOSEPH P. FOLGER, THE PROMISE OF MEDIA- TION: THE TRANSFORMATIVE APPROACH TO CONFLICT 46, 217-18 (Rev. ed. 2005) (describing transformative theory as based on assisting parties to transform their conflict interaction through empowerment and recognition, rather than focusing on achieving settlement); Robert A. Baruch Bush & Sally Ganong Pope, Changing the Quality of Conflict Interaction: The Principlesand Practiceof Transformative Media- tion, 3 PEPP. Disp. RESOL. L.J. 67, 77 (2002) ("In the transformative mediation process, parties can recapture their sense of competence and connection, reverse the negative conflict cycle, re-establish a constructive (or at least neutral) interaction and move forward on a positive footing, with the mediator's help"); Joseph Folger, Har- mony and Transformative Mediation Practice: Sustaining Ideological Differences in Purpose and Practice, 84 N.D. L. Rev. 823, 844-48 (2008) (articulating four types of transformative mediation techniques that "characterize the essential elements of transformative interventions": allowing parties to control process, mediator's maintenance of non-directive role, encouraging parties' expression and examination of differences, and supporting parties' transformations toward enlightenment and self- empowerment); Tina Nabatchi & Lisa Blomgren Bingham, Transformative Mediation in the USPS Redress Program: Observations of ADR Specialists, 18 HOFSTRA LAB. & EMP. L.J. 399, 401-02 (2001) (examining transformative mediation in employment setting as viable alternative to traditional adversarial-based process and as vehicle for parties to seize greater control over their own conflicts and learn how to effectively manage future conflicts). Note, however, that some commentators have speculated that the definition of mediation included in enabling legislation may exclude the transformative model of mediation because it is not focused on settlement. See Si- meon H. Baum, Lessons from Russian Mediators, in CONTEMPORARY ISSUES 2010, supra note 17, at 449-51. 143. David Burt, DuPont Corporate Counsel, has observed approvingly that mediation "has an intimacy foreign to adversarial methods. There is less separating the leaders and lawyers from their counterparty at mediation than in any other phase of formalized dispute resolution." David H. Burt, Inside Counsel as Sophisticated Users, in CONTEMPORARY ISSUES 2010, supra note 17, at 418, 420. Spring 2013] Bilateral Investment Treaty Arbitration even the parties) beforehand to learn about the dynamics of the dispute and interests of the parties that will be relevant in customizing the process.' 44 Mediators using most of these models may also use caucuses (or ex parte meetings), but primarily to supplement or assist the productivity of the parties'joint negotiations. Mediators may use caucuses for a variety of purposes-e.g., providing parties with the opportunity to "cool off' or express themselves on sensitive topics, offering an empathetic ear, helping parties consider how they can participate or negotiate more effectively in the joint session, encouraging parties to discuss the weaknesses of their position, or helping parties consider the consequences of the solution that appears to be the most likely candidate for settlement. All of these interventions will be designed to enhance the parties' autonomy and control of the negotiation. There is another quite different cluster of mediation models, however. These are described as "evaluative,"145 "directive,"146 fo- 14s cused on bargaining 147 and facilitating distributive outcomes.

144. See Riskin & Welsh, supra note 113, at 928; ABA SECTION OF DISPUTE RESO- LUTION, TASK FORCE ON IMPROVING MEDIATION QUALITY: FINAL REPORT APRIL 2006 - FEBRUARY 2008 10 (Am. Bar Ass'n Section of Disp. Resol. Ed., 2008) [hereinafter ABA TASK FORCE]; FRIEDMAN & HIMMELSTEIN, supra note 137, at 17-40 (2009) (describing pre-mediation discussion of dynamics involved in a dispute, which assisted the parties in selecting the appropriate model of mediation); see also John M. Barkett, Avoiding the Costs of InternationalCommercial Arbitration: Is Mediation the Solution?, in CON- TEMPORARY ISSUES 2010, supra note 17, at 367, 380-81 (describing the ICDR mediation rules, which provide the mediator with right to require shared or confidential pre-mediation submissions, as well as the CPR European Mediation Rules, which provide for confidential pre-mediation submission and mediator's authority to require exchange of written statements and other materials); Susskind, Consensus Building and ADR, supra note 65, at 361, 364 (describing the convening and conflict assessment functions in facility siting cases, and mediators' role in recommending stakeholders to be included in the process as well as whether the process should move to a "problem-solving stage"). 145. See Riskin, supra note 135, at 44-45 (noting that although evaluative approach may make it easier for parties to reach resolution because evaluative mediator provides recommendations and assessments, thereby removing some of parties' decision-making burdens, the mediator's evaluations may impede parties' ability to appreciate their own and each other's positions and make the process more antagonistic). 146. See Riskin, supra note 136, at 30 (defining term directive as "almost any conduct by which the mediator directs the mediation process, or the participants, toward a particular procedure or perspective or outcome") (emphasis in original). 147. See Silbey & Merry, supra note 138, at 19-20 (describing bargaining style mediation as rigid process driven by mediators who "claim authority as professionals with expertise in process, law, and the court system" and blatantly control proceedings, "ignoring emotional demands and concentrating on demands that can be traded off," often preferring caucuses over direct party communication and advising parties of "the benefits of a settlement of any kind"). Harvard Negotiation Law Review [Vol. 18:71

Again, mediators are likely to engage in substantial pre-mediation work. In this second set of models, though, the mediators play a more central role, and the focus is on the use of caucus rather than joint sessions. 149 If the mediator begins with a joint session, he will listen to initial presentations, generally made by the lawyers, and then quickly shift the focus to analysis of the case and his or her provision of indirect or direct advice to the parties and their lawyers. Most, if not all, of this analysis and advice will occur in caucus. Notice that this cluster of models of mediation begins to sound much like the behavior of a conciliator shuttling back and forth between the parties and "reality testing" with questions, information, and even proposals, 150 or an ICC Review Board providing informal assistance to the parties, 151 or even an arbitrator in a non-binding procedure. 152 Besides the differences in mediator role and "look" among these different models, they also vary in their substantive focus. Mediation is generally described as an interest-based process while arbitration is described as rights-based. Thus, mediation is automatically understood as encouraging the discussion of extra-legal interests to permit the development of "integrative solutions.' 53 As this discussion of the different models of mediation should make apparent, however,

148. RISKIN ET AL., supra note 127, at 311-24; CARRIE J. MENKEL-MEADOW, ET AL., MEDIATION: PRACTICE, POLICY, AND ETHics 279-86 (2006). 149. Usually, the focus is also on law-related issues, though this issue orientation is not inevitable. See Riskin, supra note 135, at 24; Riskin, supra note 136, at 23; See Riskin & Welsh, supra note 113, at 928. 150. See also Barkett, supra note 144, at 387-88 (describing the authorization of evaluative interventions, including case evaluations and settlement recommendations, in the ICDR Mediation Rules, CPR European Mediation Procedure, and UNCI- TRAL Conciliation Rules). 151. See Wolrich, supra note 23, at 468 (describing a process that involves many caucuses, to permit the members of the Dispute Board to learn enough information to enable them to provide informed advice to the parties on the issues in dispute). 152. See Salacuse, supra note 9, at 35 (equating conciliation and non-binding arbitration); see also Tom R. Tyler, Citizen Discontent with Legal Procedures: A Social Science Perspective on Civil Procedure Reform, 45 Am. J. COMP. L. 871, 877 (1997) (equating courts' non-binding arbitration programs with mediation). Sometimes, the definition of conciliation seems as contested as that of mediation. See Complementary Use, supra note 9, at 14-21; CHRISTIAN BUHRING-UHLE ET AL., ARBITRATION AND MEDI- ATION IN INTERNATIONAL BusINEss 273 (Julian Lew ed., 1st ed. 1996); see also Wolrich, supra note 23, at 473 (describing ICC's Dispute Review Board's procedures as involving a hearing, followed by the Board's recommendation; if the recommendation is not rejected by the parties, it then represents a binding contract). See e.g. S.I. Strong, Discovery Under 28 U.S.C. 1782: Distinguishing International Commercial Arbitra- tion and InternationalInvestment Arbitration, 1 STAN. J. COMPLEX LITIG. _ (forthcoming 2013). 153. See generally MENKEL-MEADOW ET AL., supra note 127, at 270-296; RiSIUN ET AL., supra note 127, 184-86 (4th ed. 2009). Spring 20131 Bilateral Investment Treaty Arbitration reality presents a more nuanced picture. 154 Much court-connected and court-oriented civil mediation in the U.S. focuses primarily on legal and litigation analysis. In the second cluster of mediation models, for example, mediators' analysis and advice generally is focused on helping the parties to be sufficiently realistic regarding their options in civil litigation (or administrative adjudication) and to guide them toward a resolution generally consistent with those options. There likely are multiple reasons for this legal and litigation-oriented focus: the mediation is occurring in the shadow of the courthouse; the lawyers generally select the mediators; 15 5 the mediators tend to be lawyers themselves, with relevant substantive expertise;1 5 6 the lawyers tend to dominate the discussion in mediation; 5 7 the parties may be unable or unwilling to engage in the sort of disclosure and discussion likely to lead to integrative solutions;' 58 and the parties may 15 9 prefer to focus on the law as a means to reach resolution. Interestingly, there are indications that sophisticated users of commercial mediation prefer mediators and mediation sessions that can uncover and use the parties' extra-legal interests and, at the same time, engage the lawyers and parties in a thoughtful discussion

154. Besides the differences in substantive focus in mediation, there are similar differences in arbitration. In some industry contexts, for example, arbitration can be explicitly interest-based rather than rights-based "Interest arbitration" is used in the labor context, to arrive at collective bargaining agreements. See RICHARD C. KEARNEY WITH DAVID G. CARNEVALE, LABOR RELATIONS IN THE PUBLIC SECTOR 264-65 (3d ed. 2001); MENKEL-MEADOW ET AL., supra note 127, at 400-01. 155. See Bobbi McAdoo, A Report to the Minnesota Supreme Court: The Impact of Rule 114 on Civil Litigation Practice in Minnesota, 25 HAMLINE L. REV. 401, 434 (2002); Bobbi McAdoo & Art Hinshaw, The Challenge of InstitutionalizingAlternative Dispute Resolution: Attorney Perspectives on the Effect of Rule 17 on Civil Litigationin Missouri, 67 Mo. L. REV. 473, 524 (2002). 156. See McAdoo, supra note 155, at 434. 157. See Nancy A. Welsh, Making Deals in Court-Connected Mediation: What's Justice Got To Do With It?, 79 WASH. U. L.Q. 787, 802-03 (2001) (citing to studies showing lawyers' domination of mediation sessions); Roselle L. Wissler, Representa- tion in Mediation: What We Know from Empirical Research, 37 FoRDHAM URB. L.J. 419, 444-45, 449-50 (2010) (comparing parties' and lawyers' level of participation in mediation sessions, as well as parties' "sense of voice"). 158. See Nancy A. Welsh, You've Got Your Mother's Laugh: What Bankruptcy Me- diation Can Learn from the Her/History of Divorce and Child Custody Mediation, 17 AM. BANKE. INST. L. REV. 427 (2009). 159. See id. at 438 (describing why so much mediation is focused on analysis of law and litigation); Welsh, Making Deals, supra note 83, at 805-06; Welsh, Thinning Vi- sion, supra note 122, at 26-27; Robert Ackerman, Disputing Together: Conflict Resolu- tion and the Search for Community, 18 OHIO ST. J. ON Disp. RESOL. 27, 55 (2002) (urging that parties who seek access to the courts also seek the application of legal norms). 114 HarvardNegotiation Law Review [Vol. 18:71 of the relevant law and litigation realities. 160 Mediation certainly has the potential to house both functions, without one dominating or marginalizing the other. Experience in the U.S. generally suggests that the most effective court-connected mediators are those who can combine elements of all of the mediation models described supra, with mediators thoroughly preparing themselves and facilitating the preparation of the disputants and their lawyers, seeking to understand important interests and develop trust, listening carefully and effectively, asking parties to explore or justify their assumptions and predictions regarding legal outcomes, carefully challenging unrealistic assumptions and helping parties to be more realistic, offering face- saving strategies, and assisting disputants and lawyers to develop responsive, realistic solutions. 161 In other words, this presumptively interest-based process is most likely to be helpful when it includes both legal analysis and probing for interests. Note, however, that the process still presumes that no agreement will occur without the parties' voluntary assent.

160. See Riskin & Welsh, supra note 113, at 928; Welsh, Stepping Back, supra note 81, at 573, 671; ABA TASK FORCE, supra note 144, at 12-13, 17, 19 (recommending further study of analytical techniques in mediation); Kenneth Kressel, The Strategic Style in Mediation, 24 CONFLICT RESOL. Q. 251, 252-3 (2007); Cheryl Picard, Exploring Integrative Framework for UnderstandingMediation 21 CONFLICT RESOL. Q. 295, 309 (2004) (arguing for a vision of mediation that integrates various theories). 161. See Riskin & Welsh, supra note 113, at 928; Welsh, Stepping Back, supra note 81, at 573, 671; ABA SECTION OF DISPUTE RESOLUTION, supra note 144, at 12-13, 17, 19 (noting the importance of mediator's involvement in preparation; also recommending further study of analytical techniques in mediation); Kressel, supra note 160, at 252-53; Picard, supra note 160, at 307-9 (arguing for a vision of mediation that integrates various theories); A. Timothy Martin, InternationalMediation: An Evolv- ing Market, in CONTEMPORARY ISSUES 2010, supra note 17, at 410, 410 (describing results of CEDR Audit, including showing that one of the primary factors facilitating settlement in mediation is parties' preparation). This article does not address the use of a single neutral to facilitate negotiation and, if necessary, serve as arbitrator. Re- gardless of whether the settlement phase of such a process is called mediation or is simply incorporated into arbitration, it will be essential to consider the potential negative effects of ex parte proceedings during the settlement phase. See Welsh, Stien- stra & McAdoo, supra note 83 (discussing potential perceptions of coercion, partiality and lack of even-handedness when the presiding judge also facilitates settlement sessions). Some organizations have adopted rules to prevent or mitigate these effects. See, e.g., Ellen E. Deason, Combinations of Mediation and Arbitration: The Challenge of Judicial Review, 5 PENN ST. Y.B. ON ARB. & MEDIATION (forthcoming 2013) (describing the Hong Kong Arbitration Ordinance, which permits med-arb with a single neutral, but requires the parties' consent and, if the parties proceed to arbitration, the neutral's disclosure of any confidential information learned during mediation that the neutral considers "material to the arbitral proceedings"); see also CEDR Rules for Facilitation of Settlement in International Arbitration, Article 5.2.1-5.2.2 ("The Arbi- tral Tribunal shall not... meet with any Party without all other Parties being present; or obtain information from any Party which is not shared with the other Parties."). Spring 2013] Bilateral Investment Treaty Arbitration 115

From a dispute system design perspective, mediation has the potential to be unique-and meaningfully different from arbitration- because it offers a lower-cost alternative that encourages communication and negotiation between the parties and explicitly includes consideration of parties' interests as a key element in resolving disputes. As the discussion of different models of mediation reveals, however, mediation's ability to realize this potential will depend upon the particular model that is integrated. Importantly, even this Article's recounting of the important variations among mediation models is likely to be incomplete, because it reflects the cultural expectations of a common law, adversarial system. Imagine that "mediation" is the name that has been given to parties' informal meetings with judges, or to disputants' informal meetings with village or community elders, in which the judges or elders will have the power to make decisions that will be enforced by the law or community. 162 It will be difficult to discern (or believe in) a conceptual distinction, in terms of outcome control, between this form of "mediation" and binding arbitration. Similarly, when judicial "adjudication" looks like a meeting, as it does in some civil law and inquisitorial systems, it may be difficult to explain how the "look" of the "adjudicative" process of arbitration should be expected to differ from mediation in terms of formality and procedural structure. The existence of these sorts of legal-cultural idiosyncrasies counsels that if mediation is added to the investment context, rules should clarify the models that may be used, including their locus of outcome control, 16 3 their "look," and the issues and norms that will be considered.164

162. See Elsing, supra note 59, at 314-17 (expressing preference for proactive arbitrators reflecting a civil law approach, also referenced as "town elder" or "arbitrator" with "back bone"); see also Baum, supra note 142, at 443 (referencing Russian parties' use of "vigilante mediation" or autoritat, with mob boss serving as mediator). 163. See Welsh, Hollow Promise, supra note 122, at 179 (exploring the nuances of parties' control over outcomes). 164. There may also be a need to overcome prejudices against the approaches that characterize foreign litigation systems, perhaps especially the systems of dominant nations. See, e.g., A. Timothy Martin, InternationalMediation: An Evolving Market, in CONTEMPORARY IssuEs 2010, supra note 17, at 415, 415 (observing that there may be suspicion of a process originating in the U.S. or U.K.). See also STEPHEN W. SCHILL, THE MULTILATERALIZATION OF INTERNATIONAL INVESTMENT LAw 6-8 (2009) (explaining many former colonies' suspicion and rejection of the multilateral investment treaty schemes proposed by the developed countries that had been colonial pow- ers), Mary Helen Mourra, The Conflicts and Controversies in Latin American Treaty- Based Disputes, in LATIN AMERICAN INVESTMENT TREATY ARBITRATION: THE CONTRO- VERSIES AND CONFLICTS 5-10, 15-17 (Thomas E. Carbonneau, ed. 2008) (regarding colonial hesitation and development of the Calvo Doctrine). Harvard Negotiation Law Review [Vol. 18:71

3. Recommended "Default"Model of Mediation for the Investor-State Context

Arguably, at least, the aim of mediation in the investment context should be enhancing parties' ability to communicate, inform, and negotiate directly with each other. After all, it will be important for the parties to maintain or improve ongoing relationships, collaborate on the implementation of any agreement, and acknowledge volatile political situations (often accompanied by difficult emotions) to enable representatives (and their constituencies) to embrace good solutions, even if they are not everyone's preferred solutions. All of these factors suggest the value of a "default" model of mediation that begins with facilitative or elicitive interventions and a focus on interests. Such a model should be preceded by careful preparation. Importantly, however, this model of mediation should also be supplemented as necessary with evaluative or directive interventions and consideration of legal rights and norms. As we have discussed supra, it is the combination of these interventions that is the hallmark of effective mediators. A process that begins facilitatively should enable the parties' "mutual consideration"' 65 of each other's perspectives and underlying needs. In other words, it should facilitate the parties' ability to engage in a procedurally just process with each other. In- vestors and states will need sufficient opportunity to speak and be heard, but also to listen to each other, reflect upon what was said, demonstrate that they have listened to each other, and also make 166 meaningful movement toward resolution. This recommendation assumes that states and investors need access to mediation because they currently have only three other procedures available to them-negotiation, conciliation, and arbitration-

165. Welsh, Stepping Back, supra note 81, at 606. 166. Id. at 620-626. But see Hollander-Blumoff, supra note 89, at 477-79. One of the authors and Professor Leonard Riskin have previously proposed a different "default" model for court-connected and court-oriented mediation. That model is presumptively law-and-litigation-focused, though it also provides for explicit consultation with the parties (or clients) so that they can elect to discuss and try to resolve issues that are broader, or extra-legal. See Riskin & Welsh, supra note 113. The default model of mediation proposed there is different from the proposal in this Article due to the differences in assumptions regarding needs and objectives. Here, there is the assumption that investors and states will need to maintain and even improve ongoing relationships, collaborate on the implementation of solutions, and find a means to acknowledge volatile political situations. For many of the non-family civil matters that are mediated in courts in the U.S., there will be no ongoing relationship between the parties and settlement of the lawsuit will involve only a one-time payment of money. Spring 2013] Bilateral Investment Treaty Arbitration to resolve their disputes. 167 The "default" mediation model that is presumptively facilitative and interest-based therefore offers something new and useful. First, of course, it provides a third party to assist the parties' negotiations; this differentiates it from negotiation. Second, its focus is on facilitating the parties' communication, information-sharing and negotiation, thus placing it within the "facilitated" category of processes, while conciliation and binding arbitration fit into the "advisory" and "imposed" categories, respectively. Finally, this model of mediation provides an explicit opportunity to identify and focus on the discussion of interests, while conciliation and arbitration presumptively focus on rights. As a "default," parties may elect to depart from this model, but they must do so explicitly and agree upon such a departure.

4. The Relationship Between the "Default"Model and States' Preventative Initiatives A few states have already expanded the continuum of options available in the investor-state context. They have established new structures and procedures designed to encourage early communication and information sharing between states and investors, thus preventing disputes or resolving them very early in their evolution. Examples include Peru's coordination and response system, including its Special Commission, and Colombia's lead agency model. 168 Peru's coordination and response system places certain obligations on any company operating in Peru if disputes arising from its operations could be governed by international arbitration guidelines, such as those contained in the ICSID Convention. 169 Peru's Ministry of Eco- nomic and Finance is responsible for collecting information regarding all parties who could be involved in such an arbitration. In this way, the Ministry can consolidate all available information regarding a

167. The parties could authorize arbitrators to engage in "interest arbitration," though, to help them arrive at a new contractual arrangement that includes terms designed to resolve past disputes. See KEARNEY WITH CARNEVALE, supra note 154, at 94-95. Of course, interest arbitration, like any adjudicative process, has the potential to produce unanticipated and unwelcome results. This is less likely to occur, however, if states and investors have confidence in the pool of arbitrators available for the arbitration process. 168. See INVESTOR-STATE DISPuTES, supra note 7, at 49, 68. 169. The goals of the system are to: 1) effectively provide information to local governments when the national government enters into a BIT of any sort; 2) provide an efficient means for the local governments to report issues with companies operating in Peru to the national government; and 3) provide a gateway for the foreign investors to have centralized contact with the national government. See INVESTOR-STATE Dis- PUTEs, supra note 7 at 69-70. 118 HarvardNegotiation Law Review [Vol. 18:71 dispute and facilitate more effective and efficient settlement negotiations. Peru credits its new system with averting 300 potential arbi- 170 tration proceedings. Similarly, Colombia's government has set up a lead agency model which makes one central agency responsible for every step of the investor relationship and arbitration process, from gathering and reporting information, transmitting information to the proper agencies within the government for analysis, and representing the government in negotiations with investors and, if necessary, in the arbitration process. Most importantly for purposes of this Article, as the governmental agency responsible for the relationship with the investor and information-gathering, the lead agency facilitates discus- sions and settlement if disputes arise between the investor and other entities within the Colombian government. These procedures may fit into the preventative, negotiated, facilitated, fact-finding or advisory categories, depending upon when and how they are implemented and the outcomes they tend to produce. 171 Indeed, and as specifically noted in Columbia, the state officials leading these efforts have the potential to serve as "quasi-mediators," meeting many of the needs that would otherwise fall to third party mediators using the presumptively facilitative or elicitive models described supra. Under these circumstances, there may be less need for the integration of a new process named "mediation." Trust, however, is the key. If investors perceive these agency quasi-mediators as biased against them or biased toward the government-thus insufficiently trustworthy, insufficiently knowledgeable, insufficiently even-handed or neutral, 72 or insufficiently open- minded-then investors are unlikely to perceive their processes or

170. INVESTOR-STATE DISPUTES, supra note 7, at 69. 171. See, e.g., Bingham, supra note 78, at 32-33 (stating reasons USPS chose not to permit evaluation by mediators); JOHN FORESTER, THE DELIBERATIVE PRACTI- TIONER: ENCOURAGING PARTICIPATORY PLANNING PROCESSES 190 (2000) (urging that intervenors' claims of neutrality frequently work to the disadvantage of the less powerful). 172. See Tyler, Rule of Law, supra note 85, at 679-80 (asserting, based on empirical evidence, that people's deference to courts and police is based on issues related to procedural justice and trust-e.g., decisions based on facts, consistent application of rules, lack of prejudice and bias, respect for people's rights and respect for them a people-not the general performance of such authorities or their decisions). See Tyler, Psychological Models, supra note 85, at 852; De Cremer, supra note 98, at 640; Markell, supra note 98, at 228-29. Spring 2013] Bilateral Investment Treaty Arbitration 119 outcomes as fair.1 73 In other words, a state-run process could be perceived as immediately subject to political and social pressures. Par- ticularly if the process's use is required, its suspected partiality will reduce its likely attractiveness and long-term effectiveness. Under these circumstances, investors are more likely to perceive a mediation process involving an outside and independent mediator as preferable in terms of its procedural justice. Even the use of an outside mediator, however, does not guarantee that investors will perceive the process as sufficiently even-handed, especially if a state agency makes repeated use of a particular outside mediator who is heavily reliant on these referrals. A foreign investor is likely to become concerned about such a mediator's ability to remain sufficiently neutral over time.174 At the very least, however, providing investors with the affordable and easily-accessible option of using a jointly-selected, independent, outside mediator may incentivize the agency to make its own processes appear, and actually be, sufficiently even- 5 handed in order to attract and maintain foreign investment. 17

173. See Welsh, Embedded Neutrals, supra note 112, at 416-27; Sean Nolon, Sec- ond Best Practices?:Addressing Mediation's Definitional Problems in Environmental Siting Disputes, 49 IDAHO L. REV. 69 (2012) (examining potential conflicts of interest for interested parties playing the role of "quasi-mediators" in the resolution of siting disputes). 174. See Welsh, Stepping Back, supra note 81 at 666 (observing that parents no- ticed special familiarity between mediator and school officials in special education mediation; also observing that mediator's inclusion on the special education mediation roster was more likely for those applicants with experience in education; experience as a parent of a child with special needs was not considered helpful); Hiram E. Chodosh, The Eighteenth Camel: Mediating Mediation Reform in India, 9 GER. L. J. 251, 280 (2008) (observing that "liln well developed systems, mediators chosen by repeat players have to counter the perception and leanings of any favoritism against single players" but also noting that "[iun newly developing processes, where trust between the parties and the mediator is critical and may be frequently based on a prior relationship, it may be harder to insist on norms of social neutrality"); see also Carrie Menkel-Meadow, Do the "Haves" Come Out Ahead in Alternative Judicial Systems?: Repeat Playersin ADR, 15 OHIO ST. J. ON Disp. RESOL. 19, 32-37, 46-47 (1999); James R. Coben, Gollum, Meet Smdagol: A Schizophrenic Rumination on Mediator Values Beyond Self Determination and Neutrality, 5 CARDOZO J. CONFLICT RESOL. 65, 77 (2004); Welsh, Contender, supra note 81, at 1168-72. 175. See, e.g., About REDRESS, USPS, Oct. 28, 2012, http://about.usps.com/what- we-are-doinglredress/about.htm [hereinafter USPS REDRESS] (describing the USPS REDRESS mediation program as an alternative to the internal EEO process); see also Maurits Barendrecht, Courts, Competition and Innovation, in TILBURG LAW SCHOOL LEGAL STUDIEs RESEARCH PAPER SERIEs No. 003, 11-12 (2012) (listing areas requiring improvement by courts that invite litigants to shop for other, more innovative, forums); Susan Sturm & Howard Gadlin, Conflict Resolution and Systemic Change, 2007 J. Disp. RESOL. 1, 42 (2007) (observing that an "institutional intermediary," unlike an outside mediator, can be "poised to integrate individual conflict resolution and systemic improvement"). HarvardNegotiation Law Review [Vol. 18:71

B. Mediation Referral Schemes As we contemplate whether and how to integrate mediation into the investment treaty context, another key question will be whether the use of mediation should be wholly voluntary or structured as a compulsory stop on the way to arbitration. Increasingly, commercial contracts in the U.S. provide for mediation as one step of several in a dispute resolution clause. 176 On occasion, one party will go to court to force the other party to mediate. 177 Courts in the U.S. handle these requests differently. Some courts enforce mediation as a contract term.178 Other courts look to the Federal Arbitration Act for guidance to enforce the mediation clause.1 79 Still other courts assess

176. See Kathleen M. Scanlon & Harpreet K. Mann, A Guide to Multi-Step Dispute Resolution Clauses, in 20 ALT. TO HIGH COST LITIG. 1, 2 (2002); Robert N. Dobbins, The Layered Dispute Resolution Clause: From Boilerplate to Business Opportunity, 1 HASTINGS Bus. L.J. 161, 166-71 (2005); CPR Model Clauses and Sample Language, CONFLICT PREVENTION AND RESOLUTION, Oct. 28, 2012, https://www.cpradr.org/Re- sources/ALLCPRArticles/tabid/265/ID/635/CPR-Model-Clauses-and-Sample-Lan- guage.aspx; S.I. Strong, Arbitration and the Republic of Colombia: Commercial, Comparative and ConstitutionalConcerns from a U.S. Perspective, 22 DUKE J. CoMP. & INT'L L. 47, 82 (2011). 177. In re Atl. Pipe, 304 F. 3d.135, 147 (1st Cir. 2002). 178. See Nancy A. Welsh, Arbitration and Beyond: Avoiding Pitfalls in Drafting Dispute Resolution Clauses in Employment Contracts, 1 J. ALT. DIrp. RESOL. EMP. 35, 41 (1999) [hereinafter Arbitration and Beyond]; Tattoo Art, Inc. v. Tat Int'l, LLC, 711 F. Supp. 2d 645, 650 (E.D. Va. 2010) ("[Tlhe mediation provision must be interpreted in accordance with Virginia contract principles. The plain language of the mediation provision unambiguously shows that the parties elected not to be subject to this Court's jurisdiction, at least with respect to any dispute stemming from the Licensing Agreement, until one of the parties either requests mediation and that request is denied or mediation commences and fails."); Vanum Constr. Co. v. Magnum Block, LLC, 245 P.3d 1069, 1075 (Kan. Ct. App. 2010) ("(1) mediation clauses should generally be construed to require a plaintiff to pursue mediation before filing a claim, even in the absence of explicit language requiring that mediation precede litigation; (2) the purpose of a mediation clause is fully served only if mediation precedes litigation; and (3) if a plaintiff fails to comply with a mediation clause before filing a claim in court, the plaintiffs claim may be dismissed."). 179. See Arbitration and Beyond, supra note 178, at 41; Sekisui Ta Indus., LLC v. Quality Tape Supply, Inc., No. DKC-08-2634, 2009 U.S. Dist. LEXIS 61983, at *11-12 (D. Md. Jul. 17, 2009) ("Unlike the Eleventh Circuit, the Fourth Circuit has not concluded that mediation is outside the scope of the FAA. In fact, as indicated above, Maryland state and federal courts have suggested that mediation does fall under the provisions of the FAA."); Advanced Bodycare Solutions, LLC v. Thione Int'l, Inc., 524 F.3d 1235, 1239 (11th Cir. 2008) ("Thus, when there is a dispute about whether any particular dispute resolution method chosen in a contract is FAA arbitration, we will look for the "common incidents" of "classic arbitration," including (i) an independent adjudicator, (ii) who applies substantive legal standards (i.e. the parties' agreement and background contract law), (iii) considers evidence and argument (however formally or informally) from each party, and (iv) renders a decision that purports to resolve the rights and duties of the parties, typically by awarding damages or equitable relief."); Gate Precast Co. v. Kenwood Towne Place, LLC, No. 1:09-CV-00113, 2009 Spring 2013] Bilateral Investment Treaty Arbitration whether mediation is likely to be fruitful, particularly in light of one party's apparent reluctance to mediate.180 Courts using this last approach generally have been less likely to enforce mediation clauses, in light of one of the parties' opposition to the process, in much the same way that some arbitral panels have not enforced the cooling off periods that precede submission of a dispute to investment treaty arbitration.1 8 .

U.S. Dist. LEXIS 100579, at *12-13 (S.D. Ohio Oct. 27, 2009) (The explicit terms of 9 U.S.C. § 3 refer to "arbitration" as opposed to the broader term "dispute resolutions procedure." Should any Act govern this question, it is rather the Uniform Mediation Act, which as Plaintiff notes, contains no mandatory enforcement provision. Media- tion is not the same as arbitration, due to its non-binding nature."). 180. See Arbitration and Beyond, supra note 178, at 41; see also, Peter Salem, The Emergence of Triage in Family Court Services: The Beginning of the End for Mandatory Mediation? 47 FAm. CT. REV. 371, 371-88 (2009) (comparing the tiered versus triage approach in ordering parties to family mediation, and urging adoption of triage approach to make use of mediation more likely to be fruitful), cited in Welsh, supra note 158, at 454 n. 175; Tattoo Art, Inc. 711 F. Supp. 2d at 652 ("Arguably, dismissing the instant lawsuit may ultimately prove inefficient and futile because the parties are not required to actually resolve the dispute through mediation. As men- tioned above, the parties are merely required to request mediation prior to initiating litigation.") 181. See Arbitration and Beyond, supra note 178, at 41; Christoph Schreuer, Calvo's Grandchildren: The Return of Local Remedies in Investment Arbitration, 4 LAw & PRAc. OF INT'L CTS. & TRIBUNALS 1, 3-5 (2005) (Discussing the non-application of treaty provisions requiring the use of local courts for a certain period of time before seeking ICSID jurisdiction. "The shorter periods look quite unrealistic and make one wonder if the drafters seriously expected that a settlement might be achieved in this way. Even the longer periods are somewhat optimistic and look more like a cooling off period than a serious attempt to settle the dispute domestically."); Siemans A.G. v. The Argentine Republic, ICSID Case No. ARB/02/8, Decision on Jurisdiction, I 32- 110 (Aug.3, 2004) https://icsid.worldbank.org/ICSID/FrontServlet?requestType= CasesRH&actionVal=showDoc&docId=DC5OSEn&caseId=C7 (Rejecting the application of an 18-month period for resort to national courts due to the petitioner's invoca- tion of a Most Favoured Nation clause in the applicable Bilateral Investment Treaty, claiming that such a condition is not imposed by Argentina under Chile-Argentina BIT, despite its presence in the Germany-Argentina BIT.); Emilio Agustfn Maffezini v. The Kingdom of Spain, ICSID Case No. ARB/97/7, Decision of the Tribunal on Ob- jections to Jurisdiction, 991 19-64 (Jan.25, 2000) https://icsid.worldbank.org/ICSID/ FrontServlet?requestType=CasesRH&actionVal=showDoc&docId=DC565_En&case Id=C163 (Rejecting Spain's challenge to jurisdiction on the basis that Spain did not condition its consent to ICSID jurisdiction on the requirement that the investor comply with the six-month waiting period and subsequent 18-month period of local jurisdiction before submitting the case to an ICSID tribunal. Because the investor invoked the treaty's Most Favoured Nation clause, the investor could operate under the conditions of more favorable treaties signed by Spain that do not have such waiting periods.). But see Richard Deutsch, An 1CSID Tribunal Denies Jurisdictionfor Failureto Satisfy BIT's "Cooling-Off"Period: Further Evidence of a Sea Change in Investor-State Arbitration or a Meaningless Ripple? 33 HoUs. J. IN''L L. 589, 592-93 (2011) (citing the Tribunal's decision in Murphy Exploration and Production Company Interna- tional v. Republic of Ecuador, ICSID Case No. ARB/08/4, Award on Jurisdiction, 1 5, Harvard Negotiation Law Review [Vol. 18:71

Meanwhile, many courts in the U.S., especially state courts, have adopted compulsory mediation programs for the resolution of non- family civil cases (e.g., contract, property damage, personal injury, employment, etc.).18 2 In general, such adoptions occurred after courts found that their purely voluntary programs were receiving little usage 83 while compulsory mediation enjoyed very substantial settlement rates, even if they were not quite as high as those produced by parties' purely voluntary use of the process.18 4 There are

10-12 (Dec. 15, 2010), in which the Tribunal denied jurisdiction because the investor did not satisfy the negotiation requirements of the Ecuador-U.S. BIT). 182. See McAdoo & Welsh, supra note 117 (generally describing state and federal courts adoption of mediation and other ADR programs for the resolution of civil non- family disputes). See also Donna Stienstra, ADR in the Federal District Courts: An Initial Report (Federal Judicial Center, 2011) available at http://www.fc.gov/publid pdf.nsf/lookup/adr20ll.pdf/$file/adr2Oll.pdf. Federal and state agencies in the U.S. have also adopted variations of mandatory mediation. And outside the U.S., some countries have made (or attempted to make) mediation a condition precedent to filing lawsuits. Italy is a prime example of this. See Nancy A. Welsh, The Importance of Context in Comparing the Worldwide Institutionalizationof Court-ConnectedMedia- tion, in ADR IN BUSINESS: PRACTICE AND ISSUES ACROSS COUNTRIES AND CULTURES, VOLUME II 119 (Arnold Ingen-Housz, ed., 2011); Jacqueline Nolan-Haley, Is Europe Headed Down the Primrose Path with Mandatory Mediation?, 37 N.C. J. INT'L L. & COM. REG. 981, 998-1006 (2012) (describing the experiences of other nations in mandating mediation). 183. See Roselle L. Wissler, Barriers to Attorneys' Discussion and Use of ADR, 19 OHIO ST. J. ON Disp. RESOL. 459, 462 n. 7 (2004); Richard C. Reuben, Constitutional Gravity: A Unitary Theory of Alternative Dispute Resolution and Public Civil Justice, 47 UCLA L. REV. 949, 981-83 (2000) (research suggesting low voluntary usage of ADR). Low usage rates are not inevitable. Some courts have been quite successful in promoting voluntary use of their mediation programs. See, e.g., McAdoo & Welsh, supra note 117, at 15 (regarding voluntary mediation program of the Federal District Court of the Western District of Michigan). 184. See Wissler, Court-Connected Mediation, supra note 90, at 697 (reporting that a slight majority of studies found no difference in settlement rates between mandatory and voluntary use, but some found greater likelihood of settlement with voluntary use); Heather Anderson & Ron Pi, Judicial Council of California, Adminis- trative Office of the Courts, Evaluationof the Early Mediation Pilot Programs 37-38 (2004) (reporting that in pilot study of five court-annexed civil mediation programs in California, mediated settlement rate was higher for voluntary than mandatory programs but also noting that "the mediation referral and settlement rates in most programs are actually quite similar to each other" and urging that mandatory mediation may achieve higher resolution if party preferences are sought and considered, while voluntary mediation may achieve higher use if courts encourage use of process and offer a financial incentive; settlement rate of 49% in Los Angeles was considerably lower than the overall average of 58%). See also Jack J. Coe, Jr., Settlement of Inves- tor-State Disputes through Mediation - PreliminaryRemarks on Processes, Problems and Prospects, in ENFORCEMENT OF ARBITRAL AwARDS AGAINST SOVEREIGNS 73, 75 n. 13 (Doak Bishop ed., 2009) (listing several studies regarding commercial mediation reporting settlement rates in excess of 80%); Martin, supra note 161, at 409 (reporting the settlement rates of mediations conducted by CEDR (89% settling at or within a short time of mediation), ICDR (85% within a few weeks or at mediation), ICC (80% if Spring 20131 Bilateral Investment Treaty Arbitration obvious parallels to the current situation in the investment treaty context. Even though some treaties explicitly invite the use of negotiation and other consensual processes before submitting a dispute to arbitration-and parties settle-there is actually little use of media- 8 5 tion or conciliation.1

1. Scope of Compulsory Mediation But "mandating mediation" can mean many-and very different-things. 8 6 First, courts' mediation mandate may apply to many cases, or only a few. Some courts, for example, order all cases into mediation. Others use a substantive screen for their compulsory programs, identifying only particular types of lawsuits to go to mediation, i.e., usually lawsuits that seem most likely to involve important extra-legal issues that cannot be fully resolved by a court. Other courts may require all civil lawsuits to go to mediation, but then will exempt particular types of cases. All of these represent "categorical"-and institutionally-mandated-referrals to mediation. On the other hand, some courts provide individual judges or court administrators with the discretion to order mediation on an ad hoc basis. These are described as "discretionary" referrals.' 8 7 Research has shown, however, that if judges utilize referrals frequently enough, repeat players are likely to perceive these discretionary referrals as de facto categorical referrals to mediation. 188 Despite their differences file transferred to mediation)); James A. Wall & Timothy C. Dunne, Mediation Re- search:A Current Review, 28 NEGOT. J. 217 (2012) (summarizing mediation research results, including those showing settlement rates of 60% to 80% in studies conducted before 1990 through the present); Lisa A. Kloppenberg, Implementation of Court-An- nexed Environmental Mediation: The District of Oregon Pilot Project, 17 OHIO ST. J. ON Disp. RESOL. 559, 581-82 (2002) (reporting on case management of environmental cases and outcomes). But see Burt, supra note 61, at 418 n. 1, 424 (describing court- ordered mediation in the U.S. as "nearly worthless" and "oxymoronic"). 185. See INVESTOR-STATE DISPUTES, supra note 7, at xxi-xxvii; INVESTOR-STATE DISPUTES II, supra note 1 at xv; see also U.S. DEP'T Of STATE, MODEL BIT (2004), availableat http://www.state.gov/documents/organization/l17601.pdf (referencing the potential to use consensual processes before going to arbitration). But see Deutsch, supra note 181, at 594, 603. 186. McAdoo & Welsh, supra note 117, at 10-11 (describing different mandatory schemes in state and federal courts in the U.S.). 187. Frank Sander, et al., Judicial (Mis) Use ofADR? A Debate, 27 U. TOL. L. REV. 885, 885-86 (1996), cited in Dorcas Quek, Mandatory Mediation: An Oxymoron? Ex- amining the Feasibility of Implementing A Court-Mandated Mediation Program, 11 CARDOZO J. CONFLICT RESOL. 479, 480-81 (2010); see also McAdoo, supra note 111, at 358-60 (describing approaches used by various state appellate court administrators in screening to determine whether to order parties into appellate mediation). 188. See McAdoo & Welsh, supra note 63, at 408 (noting that Minnesota judges' "routine ordering of ADR was a key factor motivating attorneys to select mediation"); Harvard Negotiation Law Review [Vol. 18:71 in the scope of the cases to which they apply, all of these options represent compulsory referral schemes.

2. Degree of Intrusiveness of Compulsory Elements Courts in the U.S. have also experimented with the level of intrusiveness of the obligations being imposed upon parties. In some courts, parties are required only to consider the use of mediation (and other dispute resolution options) on their own and submit a document to the court that responds to the court's questions regarding the appropriateness of the process.18 9 Other courts require parties to explore the potential use of mediation by attending a case conference at which mediation will be discussed. Sometimes, these conferences transform into initial mediation sessions. Still other courts explicitly require the parties to begin mediation by attending an initial orientation or first mediation session. 190 After the parties have completed this obligation and thus have had the opportunity to develop their perceptions of the mediation process and the particular mediator's capacities, they may determine whether they wish to continue with the process. Last, of course, many courts in the U.S. require the parties to participate in an entire mediation process. In general, though,

McAdoo & Hinshaw, supra note 155, at 506, 537 (2002) (reporting that one third of lawyers not using ADR made that choice because judges were not ordering or encouraging it; therefore encouraging judges to be more proactive); see also Bobbi McAdoo, All Rise, the Court is in Session: What Judges Say About Court-ConnectedMediation, 22 OHIO ST. J. ON Disp. RESOL. 377, 393-404 (2007) (reporting research regarding factors motivating judges to order the use of mediation). 189. See Bobbi McAdoo & Nancy A. Welsh, Does ADR Really Have A Place on the Lawyer's Philosophical Map?, 18 HAMLINE J. PuB. L. & POL'Y 376, 380-81 (1997) (describing mandatory consideration of ADR); Nancy A. Welsh & Bobbi McAdoo, Al- ternative Dispute Resolution (ADR) in Minnesota - An Update on Rule 114, in COURT- ANNEXED MEDIATION: CRITICAL PERSPECTIVES ON SELECTED STATE AND FEDERAL PRO- GRAMS 203, 206-207 (Edward Bergman & John Bickerman, eds., 1998) (describing mandatory consideration under Rule 114); see also Nancy A. Welsh, The Future of Mediation: Court-ConnectedMediation in the U.S. and the Netherlands Compared, 1 FORUM VOOR CONFLICT MGMT. 19, 21 (2007) (noting some "Dutch judges in the pilot sites ... distributed a 'self test' to all parties to encourage them to consider whether mediation might be appropriate for their case."); Baum, supra note 142, at 356-59 (describing Russian legislation which mandates mediation, but also includes a 30-day default provision, which allows parties to avoid mediation by failing to pursue it. This essentially guts the compulsory element of the statute; the default provision does not apply to judicial or arbitral facilitation of settlement). 190. See Coe, supra note 22, at 350 (proposing settlement attempt as condition precedent for claim administrability and that one session with mediator would represent prima facie evidence of compliance with such requirement); Ulicny, supra note 64, at 480 (observing that even when ADR is mandatory by contract, the ICC's ADR Rules permit the parties to terminate ADR proceedings after an initial meeting with the neutral). Spring 2013] Bilateral Investment Treaty Arbitration courts do not specify how long the mediation must last or how many sessions the parties must attend. The mediation, therefore, will last only as long as the parties wish to continue. In some other countries, parties' participation in mediation may not be mandated per se, 19 1 but use of certain other mechanisms can strongly incentivize parties and influence their cost-benefit analyses regarding the utility of mediation. In the U.K. and Hong Kong, for example, courts may refuse to shift litigation-related costs if they 192 judge that a party was unreasonable in its refusal to mediate. Some nations, such as the Netherlands, provide funding to support parties' participation in mediation. 193 Other nations may condition, or appear to condition, parties' access to legal aid funds upon their participation in mediation. 194 Still other nations have created incentives for the parties' lawyers to refer cases to mediation.' 95 Professor

191. See Halsey v. Milton Keynes General NHS Trust, [2004] EWCA 3006 Civ 576 (opposing compulsory mediation but permitting the award of costs against a party who refused to mediate). The Italian government adopted compulsory mediation, pursuant to Legislative Decree no. 28 (2010). See Matthew Rushton, Mandatory Me- diation Under Threat in Italy, JAMS ADR Blog (Oct. 30, 2012), http://jamsadr- blog.com/2012/10/30/mandatory-mediation-under-threat-in-italy/. On October 24, 2012, however, Italy's Constitutional Court invalidated the compulsory mediation program. Id.; Corte Cost., 24 ottobre 2012, n.272, Giur it. 2012 (It.). But see Rosalba Alassini v. Telecom Italia SpA, 2005 O.J. (C 134) 3 (holding that compulsory mediation does not violate Article 6 of the European Convention on Human Rights); see also Jacqueline Nolan-Haley, Mediation Exceptionality, 78 FORDHAM L. REV. 1237, 1258 (2009) (describing evolving British jurisprudence on compulsory mediation and imposition of costs). 192. See Nolan-Haley, Mediation Exceptionality, supra note 191; see also Nolan- Haley, supra note 182, at 999-1005 (describing the evolution of case law in England regarding the imposition of costs for unreasonable refusal to participate in mediation, as well as Italian law that may result in imposition of costs for refusal to accept "mediator's proposal"); Quek, supra note 187, at 503 (describing cost-shifting in Hong Kong). 193. See Welsh, supra note 182, at 121 (observing that when the Netherlands established its court-connected mediation program, "the government provided a mediation subsidy to those who elected the process pursuant to court invitation or the advice of a legal advice helpdesk"). See also Nolan-Haley, supra note 182, at 1005-06 (observing that Bulgaria provides a partial refund of court filing fees to parties who settle in mediation, while Romania provides a full refund of such fees). The ICC also incentivizes the use of mediation by crediting half of the administrative ADR toward the subsequent arbitration, if the mediation fails to produce a settlement. See Ulicny, supra note 64, at 481. 194. See Welsh, supra note 182, at 122. 195. See Robert Dingwall, Divorce Mediation: Should We Change Our Mind? 32 J. Soc. WELFARE & FAm. L. 107, 111 (2010) (urging that by treating mediation referrals as a performance indicator, LSC has made an "implied threat that solicitors who failed to achieve LSC targets would find their franchise brought into question"); see also Pilot InsuranceADR Project Results in Many Settlements, 7 ALTS. TO HIGH COST LITIG. 37, 40 (1989) (reporting results of pilot project involving several Connecticut Harvard Negotiation Law Review [Vol. 18:71

Jack Coe has proposed the creation of a different and positive set of incentives to use mediation in the investor-state context by having a mediator in attendance throughout arbitration sessions and available at any time (or at specific points) to facilitate the parties' discussion 96 of settlement. 1

3. Parties'Input Within Compulsory Systems

Even when courts have adopted a straightforward compulsory mediation program, they may still give serious consideration to parties' input in determining whether or not the mediation will take place. As noted supra, some courts require only that the parties consider the use of mediation and submit a document. An individual judge or court administrator may then review the submission to determine whether this particular case should be ordered into mediation. While the lawyers are likely to defer to a judge's expressed support for mediation, 197 the judge or court administrator is also likely to defer to the parties. If all of them agree that mediation is not likely to be useful, an order to participate is doubtful. On the other hand, the court is much less likely to defer to one party's reluctance if another party expresses interest in mediation. Even after a court has ordered the parties' participation in mediation, the court also may permit the parties to opt out.198 Some courts grant such

insurers and finding that use of mediation increased as a result of making mediation referrals one of the performance indicators for claims adjusters). 196. See Coe, Complementary Use, supra note 9, at 40; Coe, supra note 184, at 96- 100 (proposing "shadow mediator" and mediation as a parallel process with multiple pre-set points of engagement); see also CEDR Commission on Settlement in Interna- tional Arbitration, Final Report, 5, 7 (Nov. 2009) (providing for parallel mediation or mediation "window" during pendency of arbitration); Ulicny, supra note 64, at 481 (observing that the ICC's ADR services are available to parties during the pendency of their arbitration). This is interestingly similar to the settlement-facilitating role played by magistrate judges in some federal district courts and staff mediators in federal circuit courts, as well as the role played by community mediators in some states' small claims courts. In these courts, the community mediators are in attendance and available to parties who wish to attempt to reach a settlement through the process before appearing before the judge or referee for a binding decision. See Patrick G. Coy & Timothy Hedeen, A Stage Model of Social Movement Co-Optation:Commu- nity Mediation in the United States, 46 Soc. Q. 405, 408 (2005) (detailing the history of community mediation). 197. See McAdoo & Welsh, supra note 63, at 409. 198. See McAdoo & Welsh, supra note 117, at 16-17. Spring 2013] Bilateral Investment Treaty Arbitration 127 opt-outs on a very liberal basis. 19 9 Others require the party requesting the opt-out to make certain showings to demonstrate why the mediation would be duplicative of earlier efforts, unlikely to be 20 0 productive, or might even cause harm.

4. Mixed Compulsory and Voluntary Elements

In situations characterized by perceptions of power imbalance, some agencies and corporations have provided that their own employees or associated entities must participate in mediation if a citizen, consumer, or employee requests the process.2 0 1 Thus the mediation process is voluntary for the party generally perceived as less powerful, but compulsory for the party perceived as more powerful and that has established the program providing mediation. The U.S. Postal Service (USPS), for example, offers a mediation program to resolve workplace disputes. If an employee requests mediation to address a dispute with her supervisor, the USPS requires the supervisor to participate. 20 2 Similarly, some school districts in the U.S. provide that if a parent requests mediation to resolve a special education dispute, the school officials must participate. These approaches have had interesting effects. The USPS, for example, has found that managers have learned to listen more effectively to their employees, rather than responding too quickly with answers or orders.20 3 School officials, meanwhile, value mediators' interventions that can help clarify what parents are seeking for their children. 20 4 In other words, they also appreciate assistance with more effective listening and translation.

199. See id. 200. See id. 201. See Welsh, Contender, supra note 81, at 1171, 1176-78 (noting that marginal- ized plaintiffs have the power to force defendants to participate in litigation, parents in some school districts have the power to force school officials to participate in special education mediation, and USPS employees have the power to force their managers to participate in the REDRESS mediation program); Andrea K. Schneider & Natalie Fleury, There's No Place Like Home: Applying Dispute Systems Design Theory to Cre- ate a ForeclosureMediation System, 11 NEV. L.J. 368, 379 n. 44 (2011) (if homeowner initiated foreclosure mediation, lender was required to participate); see also Ster- nlight, supra note 118, at 85-90 (proposing that mandatory arbitration should be imposed on the company, not the "little guy"). 202. See Lisa Blomgren Bingham, supra note 79, at 113; USPS REDRESS, supra note 174 ("If REDRESS is offered and the employee requests mediation, the appropriate management official will participate in the process in good faith"). 203. See Jonathan F. Anderson & Lisa Blomgren Bingham, Upstream Effects from Mediation of Workplace Disputes: Some PreliminaryEvidence from the USPS, 48 LA. L.J. 601, 607-08 (1997). 204. See Welsh, Stepping Back, supra note 81, at 581. Harvard Negotiation Law Review [Vol. 18:71

In the investor-state context, if an investor has worked with a lead agency or pursued administrative review20 5 in compliance with a BIT's pre-arbitration requirements, and the investor then requests mediation with an independent third party mediator (or requests a case conference or initial mediation session as described supra), the BIT could compel the state or affected agency to comply with such request. One concern with this approach might be that multinational corporations, which are likely to face the same types of investment concerns around the globe, will be advantaged by having more knowledge and experience with the mediation process, the available pool of mediators, and relevant substantive issues.20 6 This concern is less compelling, of course, if the investor is neither a multinational corpo- ration nor a sophisticated repeat player in international trade. Fur- ther, in mediation, representatives of the state will not be obligated to reach an agreement unless and until they are satisfied regarding the appropriateness, value, and consequences of such agreement. In making such determinations, they would be able to consider the investor's reputation in its dealings with other similarly-situated states.

5. The Effects of Compulsory Mediation Some commentators have urged that neither legislatures nor courts should make mediation compulsory because this represents a violation of the parties' self-determination 20 7 and may have the effect of coercing settlements and reducing actual or perceived access to the

205. See INVESTOR-STATE DISPUTES, supra note 7, at 68-73 (providing a description of Peru, etc.). 206. See Edward Brunet, Reevaluating Complex Mediation Generalizations, 18 CONN. INS. L.J. 279, 281 (2011-2012) (observing that "defendant insurers are often data-rich 'repeat players' who know the score due often to access to much more information than their opponents regarding relevant judgments and settlements"); Riskin & Welsh, supra note 113, at 864-66 (describing the influence of civil litigation "repeat players" upon the mediation process); see also Rogers, supra note 14, at 357-58 (regarding repeat players' general advantages in arbitration and developing countries' lack of experience and sophistication in selecting arbitrators); Catherine A. Rogers, Regulating International Arbitrators: A Functional Approach to Developing Stan- dards of Conduct, 41 STAN. J. INT'L L. 53, 112 n. 340 (2005). 207. See Welsh, Thinning Vision, supra note 122, at 18, n. 68 (citing various commentators); John P. McCrory, Mandated Mediation of Civil Cases in State Courts:A Litigants Perspectiveon Program Model Choices, 14 OHIO ST. J. ON DiSP. RESOL. 813, 830-31 (1999). Based on the experience of the U.S., Professor Jacqueline Nolan-Haley has recently urged an end to compulsory mediation. See Jacqueline M. Nolan-Haley, Judicial Review of Mediated Settlement Agreements: Improving Mediation with Con- sent, 5 PENN ST. Y.B ON ARB. & MEDIATION - (forthcoming 2013) (proposing an end to compulsory, court-connected mediation and the requirement of informed consent to mediate). Spring 2013] Bilateral Investment Treaty Arbitration courts. 20 8 As should be obvious from the discussion supra, this concern applies primarily to the most intrusive compulsory referral schemes that require participation in the entire mediation process. Even for these types of schemes, however, some commentators have expressed support for time-limited compulsory programs in order to encourage lawyers and other repeat players to learn how to participate in the process. 20 9 Research has then demonstrated that lawyers who have experienced mediation are likely to recommend its use on a voluntary basis.210 After this provisional period of compulsion, courts have then been able to convert to voluntary programs or pro- 211 vide for easy opt-outs. The effects of compulsory referral have been generally, and perhaps surprisingly, salutary. Although settlement rates tend to be somewhat lower for compulsory mediation 212 than voluntary mediation, 213 they remain very high. The simple requirement to participate in the process, and participate knowledgeably, generates settlements. 2 14 Because compulsory mediation increases the number of cases that go to mediation, the absolute number of settlements

208. See Welsh, Place of Mediation, supra note 112, at 142. 209. Sander, et al., supra note 187, at 886; Welsh, Place of Mediation, supra note 112, at 142; McAdoo & Welsh, supra note 63, at 426. 210. See Roselle L. Wissler & Bob Dauber, Leading Horses to Water: The Impact of an ADR "Confer and Report" Rule, 26 JUST. Sys. J. 253, 263 (2005) (reporting empirical research showing that after the adoption of a court rule requiring lawyers to confer regarding ADR and report certain information to the court, lawyers also became more likely to discuss the possible use of a voluntary ADR process with their clients and voluntary use of ADR occurred more frequently). 211. See Welsh, Place of Mediation, supra note 112, at 139-41; McAdoo & Welsh, supra note 63, at 413. 212. See Wissler Court-ConnectedMediation, supra note 90, at 676, n. 137 (reporting that some research revealed no difference between voluntary and mandatory referrals, while other research showed a higher rate of settlement for voluntary use of mediation). 213. See Jeanne M. Brett et al., The Effectiveness of Mediation: An Independent Analysis of Cases Handled by Four Major Service Providers, 12 NEGOT. J. 259, 260-67 (1996) (asserting 81% settlement rate in study of 449 contract- and tort-based disputes); see also EILEEN CARROLL & KARL MACKIE, INTERNATIONAL MEDIATION - THE ART OF BUSINESS DIPLOMACY 91 (2000) (reporting Center for Dispute Resolution's settlement rates); Robert Coulson, Arbitration and Other Forms of Alternative Dispute Resolution - General Overview, in WORLDWIDE FORUM ON THE ARBITRATION OF INTEL- LECTUAL PROPERTY DISPUTES 21, 22 (1994) (reporting AAA commercial mediation settlement rates of more than 80%); James A. Wall & Timothy C. Dunne, Mediation Research: A Current Review, 28 NEGOT. J. 217, 229-30 (2012) (summarizing studies that show 60-90% settlement rate). 214. Research has long supported the settlement value of scheduling litigation "events." See Stephen J. Spurr, The Durationof Litigation, 19 LAW & POL'Y 285, 305 (1997), cited in Lisa Blomgren Bingham, et al., Dispute Resolution and the Vanishing Trial: ComparingFederal Government Litigation and ADR Outcomes, 24 OHIO ST. J. Harvard Negotiation Law Review [Vol. 18:71 achieved through mediation is higher. Depending upon the timing of the order to mediate, mediation can also produce settlements that are earlier and cost less.215 Also, in general, parties express satisfaction with these court-mandated procedures and perceive the process as procedurally fair. 216 Parties have the opportunity to craft creative, customized solutions-and sometimes they do. Of course, there are also some instances when parties are dissatisfied with their mediator or the process 217 as well as occasions when parties have refused to cooperate with courts' attendance and submission requirements designed to enhance mediation's effectiveness. 218 There is also worri- some evidence of some repeat players using the mediation process inappropriately. 21 9 Nonetheless, both the empirically-demonstrated successes and challenges experienced by domestic courts' mediation programs shed light on how to structure a mediation referral system that would fit the investor-state context, particularly if the mediation process is managed by an organization that is similar to the best

ON Disp. RESOL. 225, 256 n. 129 (2009); see also Barkett, supra note 144, at 364 (suggesting the mediation is more successful if it is preceded by some shared knowledge regarding the evidence supporting claims and defenses). 215. See DONNA STIENSTRA ET AL., FED. JUD. CTR., REPORT TO THE JUDICIAL CON- FERENCE COMMITTEE ON COURT ADMINISTRATION AND CASE MANAGEMENT: A STUDY OF THE FIVE DEMONSTRATION PROGRAMS ESTABLISHED UNDER THE CIVIL JUSTICE REFORM ACT OF 1990 19 (1997); Wissler & Dauber, supra note 210, at 270; JULIE MACFARLANE, LAW COMM'N OF CANADA, CULTURE CHANGE? COMMERCIAL LITIGATORS AND THE ONTA- RIO MANDATORY MEDIATION PROGRAM 14 (2008); see also McAdoo & Welsh, supra note 63, at 422-25 (citing Wissler, Court-ConnectedMediation supra note 90, at 694 (cites to several studies where the design of the program ensured an early mediation and, as a result, reduced discovery)); John Lande, The Movement Toward Early Case Man- agement in Courts and Private Dispute Resolution, 24 OHIO ST. J. ON DIsP. RESOL. 81, 85-86 (2008). But see Barkett, supra note 144, at 389-90 (discouraging use of mediation before arbitration commences; noting that court-connected mediation is likely to occur after the exchange of initial pleadings and some initial discovery). 216. See McAdoo & Welsh, supra note 63, at 422-425; Wissler, Court-Connected Mediation supra note 90, at 661-663; Craig A. McEwen & Roselle L. Wissler, Finding Out If It Is True: Comparing Mediation and Negotiation Through Research, 2002 J. DIsP. RESOL. 131, 140-142 (2002). 217. See Welsh, Thinning Vision, supra note 122, at 15. 218. See John Lande, Using Dispute System Design Methods to Promote Good- Faith Participationin Court-Connected Mediation Programs, 50 UCLA L. REV. 69, 86 (2002); Maureen A. Weston, Checks on ParticipantConduct in Compulsory ADR: Rec- onciling the Tension in the Need for Good-Faith Participation,Autonomy, and Confi- dentiality, 76 IND. L.J. 591, 596 (2001). 219. See Nancy A. Welsh, Musings, supra note 128, at 19. Spring 2013] Bilateral Investment Treaty Arbitration courts, i.e., sufficiently independent and neutral of the key stakehold- 220 ers, assertive regarding the value of participating in mediation 221 and responsive to the dangers of repeat player dynamics. From a dispute system design perspective, any compulsory referral scheme should maintain parties' ability to loop forward and backward among different processes. This suggests that even if the default model of mediation in the investor-state context is facilitative and interest-based mediation, supplemented by consideration of

220. Although this Article is not focusing on the important role that dispute resolution facilities can play in encouraging parties' use of mediation, it is worthwhile to note here that some courts have been very effective in helping parties to customize the process to meet their needs. See Riskin & Welsh, supra note 113, at 920-21 (discussing the approach used by mediation staff members at the U.S. District Court of the Northern District of California and the Ninth Circuit); Welsh, supra note 182, at 132 (discussing the U.S. District Court of the Northern District of California and the Ninth Circuit); McAdoo & Welsh, supra note 117, at 15 (describing one federal district court-specifically, the United States District Court for the Western District of Michi- gan-that has been very successful in persuading parties to use the court's voluntary mediation program). 221. See, e.g., Jill I. Gross, The End of Mandatory SecuritiesArbitration?, 30 PACE L. REV. 1174, 1194 (2010); Jill I. Gross & Barbara Black, When Perception Changes Reality: An Empirical Study of Investors' Views of the Fairness of Securities Arbitra- tion, 2008 J. Disp. RESOL. 349, 389 (reporting research initiated by SICA that found investors were much less likely than industry representatives to perceive procedural fairness in securities arbitration context); see also FINANCLL INDUSTRY REGULATORY AUTHORITY (FINRA), SECURITIES AND EXCHANGE COMMISSION INVESTOR ADVISORY COMMITTEE PANEL ON SECURITIES ARBITRATION - MAY 17, 2010, FINANCIAL INDUSTRY REGULATORY AUTHORITY STATEMENT ON KEY ISSUES 4 (2010), http://www.sec.gov/spot- lightlinvadvcomm/iacmeeting05l7l0-finra.pdf (reporting change in FINRA rules allowing investors to elect an all public arbitral panel, though they continue to have the option of selecting a panel consisting of two public arbitrators and one industry arbitrator); Thomas J. Stipanowich, The Arbitration FairnessIndex: Using a Public Rat- ing System to Skirt the Legal Logjam and Promote Fairer and More Effective Arbitration of Employment and Consumer Disputes, 60 KAN. L. REV. 985, 1005-07 (2012) (describing various efforts that have been undertaken by the American Arbi- tration Association, Council of Better Business Bureaus, Securities and Exchange Commission, CPR Institute for Dispute Resolution, Federal Trade Commission and other public and private entities to develop and implement protocols focused on the provision of fair procedures); Welsh, Embedded Neutrals, supra note 112, at 423-24 (complimenting FINRA and SICA for being "remarkably proactive"); Nancy A. Welsh, Mandatory Predispute Arbitration, Structural Bias, and Incentivizing Procedural Safeguards, 42 Sw. U. L. REV. 187 (2012) (describing Supreme Court's recognition of structural bias in ERISA context and proposing application of resulting test to mandatory predispute arbitration in consumer, employment and other disparate party arbitration). There is also a fairly substantial body of literature regarding the effectiveness of corporations' internal dispute systems, including the use of mediation. See, e.g., Craig A. McEwen, Managing Corporate Disputing: Overcoming Barriers to the Effective Use of Mediation for Reducing the Cost and Time of Litigation, 14 OHIO ST. J. ON Disp. RESOL. 1, 2-3 (1998). The procedural and substantive fairness of such internal systems is likely to depend, at least in part, on continued access to the courts. See Welsh, Contender, supra note 81, at 1175. HarvardNegotiation Law Review [Vol. 18:71 rights and evaluative mediator interventions as needed, the parties still should be able to elect a different consensual procedure. Parties' ability to select their preferred consensual option might also be achieved by limiting the compulsory mediation requirement to one of the less-intrusive options described supra.

6. Recommended Compulsory Elements in the Investor-State Context

Based on dispute system design and procedural justice consider- ations, this Article recommends mandating parties' participation in an initial meeting to learn about mediation and other consensual options. Even such a compulsory scheme, though, could benefit from the engagement of stakeholders and the demonstrated support of strong global actors, such as the World Bank, ICSID, and UNCTAD. The currently low incidence of voluntary usage of conciliation and mediation in the investment treaty context could suggest that the processes are not needed, or as constructed, do not respond to parties' perceived needs. In the investor-state context, however, some of the strongest proponents of mediation are states and investors-and some of the strongest opponents have not seemed to be aware of the many variations available for mediation and compulsory referral. In- stead, they have assumed that the only approach to compulsory mediation is the most intrusive one. 222 As we have outlined, this assumption is too narrow. Both dispute system design principles and procedural justice research suggest the value of engaging stakeholders directly in discussion and decision-making regarding the mediation models and referral schemes that would be most responsive in the investment treaty context. Such engagement, along with strong encouragement from important global actors such as the World Bank, is likely to produce the incentives that will encourage mediation's use. For example, if stakeholders can count on the support of the World Bank, they may embrace a period of "coerced education" as described supra to ensure that mediation will receive sufficient use to justify the simul- taneous provision of relevant education, training, and support.

222. See, e.g., Stevens, supra note 42, at 28-29 (not supportive of compulsory mediation); Reed, supra note 42, at 30 (not supportive of compulsory mediation). Spring 2013] Bilateral Investment Treaty Arbitration

C. Mediation Quality Assurance/Accountability Mechanisms223 According to dispute system design principles, once a dispute resolution system's objectives and operation have been established and stakeholders have begun electing the procedures to be used to resolve their disputes, mechanisms must be established to ensure the system's and procedures' accountability to the stakeholders. Courts and agencies in the U.S. have adopted mediation primarily in order to settle cases224 and provide parties with the opportunity to be more engaged in the resolution of their disputes.2 25 As a result, they have used three approaches to regulate the quality of mediation and hold the process accountable. Interestingly enough, the first approach focuses on regulating the parties' behavior and making them accountable to the court or agency-and each other. The second approach focuses on making the mediator accountable to the court or agency and the parties. The third approach-making courts accountable to the Congress and state legislatures (or the citizenry more generally) for the quality of their mediation programs-has been accomplished on a much more ad hoc basis. In this Article, we will do no more than introduce these different approaches. Much more could be said, and 227 has been said,226 about them.

223. See Charles H. Brower, Confronting the Truth: Sources and Magnitude of Decentralization in Investment Treaty Arbitration, in THE FUTURE OF INVESTMENT ARBITRATION 339-48 (Catherine A. Rogers & Roger P. Alford eds., 2009) (explaining current and future extreme variability of arbitral outcomes in investment arbitration). 224. See Welsh, Thinning Vision, supra note 122, at 59-60; McAdoo & Welsh, supra note 63, at 406. 225. See McAdoo & Welsh, supra note 63, at 406; McAdoo, supra note 188, at 399; Brazil, supra note 108, at 240 (2007) (exploring what it means for litigants to feel "well-served by their public institutions[,]" particularly the courts, in court-connected mediation). 226. See, e.g., Charles Pou, Jr., Assuring Excellence, or Merely Reassuring? Policy and Practice in Promoting Mediator Quality, 2004 J. DIsEp. RESOL. 303, at (2004); Charles Pou, Jr., Mediator Quality Assurance: A Report to the Maryland Mediator Quality Assurance Oversight Committee, Md. State Bar Ass'n 1 2002 (catdgorizing approaches based on "the height of the 'hurdles' that mediators must meet at the outset to engage in practice" and "the amount of 'maintenance' or development aid required to broaden their awareness and enhance skills over time"); Ellen Waldman, CredentialingApproaches: The Slow Movement Toward Skills-Based Testing Contin- ues, 8 Disp. RESOL. MAG. 13 (2001); Craig A. McEwen, Giving Meaning to Mediator Professionalism, 11 Disp. RESOL. MAG. 3 (2005); Nancy A. Welsh, Institutionalization and Professionalization,in HANDBOOK OF DISPUTE RESOLUTION, supra note 65, 494- 98. 227. This Article will not discuss parties' apparent dissatisfaction with, and disa- greement regarding, some of the terms contained in investment treaties. See Schnei- der, supra note 24 (discussing the varying interpretations used by the arbitral tribunals of the BIT emergency clauses under international law). Needless to say, 134 Harvard Negotiation Law Review [Vol. 18:71

1. Accountability of the Parties to the Court and Each Other As noted supra, procedural justice research indicates that the nature of parties' interaction with each other in mediation matters, and parties are more likely to perceive the mediation process as fair if their lawyers perceive each other as behaving cooperatively. 228 This research suggests the importance of thoughtful management of the parties' interactions, as well as the importance of lawyers' preparation of themselves and their clients for respectful, thoughtful participation.229 Other research also indicates that when disputants negotiate with each other in a manner that is perceived as procedurally just, they are more likely to maximize joint outcomes. 230 It is not clear that there is a causal relationship here, but the correlation 231 is quite striking, and provides an additional incentive for respectful, thoughtful participation. Of course, courts cannot require settlement, mutual listening, or other procedurally just behavior. What courts in the U.S. have done, however, is require the lawyers and parties to make submissions that demonstrate lawyers' preparation and consultation with their clients, 232 require the participation of lawyers and parties with authority in the mediation, and impose an obligation to mediate "in good faith."233 Thus far, courts have tended to focus on compliance with however, investors' and states' disagreements over the meaning of key terms, the intent of the drafters, and the applicability of the terms to non-signatories resemble parties' disagreements over the meaning and intent of state and federal statutes- and suggest the important role that adjudication plays in making Congress and state legislatures accountable to the citizenry. See Owen Fiss, Against Settlement, 93 YALE. L.J. 1073, 1076 (1984). 228. See Wissler & Dauber, supra note 210, at 266-68. 229. See Wissler, Court-ConnectedMediation, supra note 90, at 686-87 (reporting that clients who were more prepared by their lawyers perceived the mediation process as fairer; similarly, lawyers who were more prepared reported greater perceptions of procedural fairness). 230. See Hollander-Blumoff & Tyler, supra note 89, at 489-90; see also Welsh, Reputational Advantages, supra note 63, at 127-130 (exploring relationships among perceptions of effectiveness in negotiation, elements of a cooperative style of negotiation, interactional elements that lead to perceptions of procedural justice, and behaviors that lead to perceptions of trustworthiness). 231. It is also consistent with Morton Deutsch's "crude law of social relations" that "the characteristic processes and effects elicited by a given type of social relationship also tend to elicit that type of social relationship." Morton Deutsch, Conflict Resolu- tion: Theory and Practice, 4 POL. PSYCHOL. 431, 438 (1983). 232. See Riskin & Welsh, supra note 113, at 898 (urging courts to require lawyers to consult with their clients regarding their objectives for the mediation session, in order to permit appropriate customization). 233. See Weston supra note 218, at 596; see also Lande, supra note 218, at 78; Tania Sourdin, Good Faith, Bad Faith? Making an Effort in Dispute Resolution Spring 2013] Bilateral Investment Treaty Arbitration the specific, objective requirements referenced supra-e.g., attendance by those with settlement authority and submission of required pre-mediation documents-to shape the legal boundaries of "good faith" participation. 234 Courts regularly establish such requirements, and better practice (as well as dispute system design) suggests the value of soliciting stakeholders' input on the elements of 23 5 such an obligation.

2. Accountability of the Mediators to the Court and Parties As outlined above, the second method to ensure a quality process is to focus on the quality and skills of the mediator. Courts have adopted a wide variety of approaches to make mediators accountable to the courts for the quality of their mediation services. 23 6 Many courts try to regulate quality by establishing entry requirements. For example, numerous courts require their mediators to receive a specified number of hours of training in mediation skills; few courts, however, regulate the training programs, and research has been mixed regarding the value of mediation training.237 Some courts require their mediators to observe other "master" mediators' mediations or co-mediate with an experienced mediator before they will be

(2012), available at http://www.civiljustice.info/goodf/l (reviewing case law regarding requirement of good faith in negotiation and mediation, in Australia). 234. See also Barkett, supra note 144, at 385 (describing UNCITRAL's obligation to cooperate in good faith with conciliator); Franck, supra note 22, at 197 n.156 (discussing whether obligation to negotiate and consult before initiating arbitration requires "good faith" negotiation). 235. See Lande, supra note 114, at 127 (proposing the use of dispute system design principles for determining good faith requirements). 236. See Welsh, supra note 226, at 487; McAdoo & Welsh, supra note 117, at 22-23 (on training and qualifications). 237. See McAdoo, A Report to the Minnesota Supreme Court, supra note 155, at 434 Table 13 (2002) (reporting the following responses from surveyed lawyers to the question "What mediator qualifications are important to you?": mediator should have substantive experience in field of law related to case (84.2%); mediator should be a litigator (66.2%); mediator should be a lawyer (63.6%); mediator should have taken mediation training (42.8%); mediator has a reputation for settling cases (39.4%)); McAdoo & Hinshaw, supra note 155, at 524 Table 33 (reporting the following responses from surveyed lawyers to the question "What mediator qualifications are important to you?": mediator knows how to value a case (87%); mediator should be a litigator (83%); mediator should be a lawyer (77%); mediator knows how to help parties clarify issues (74%); mediator should have substantive experience in the field of law (69%); mediator should have taken mediation training (49%); mediator has substantial mediation experience (46%); mediator knows how to find creative solutions (35%); mediator is good at helping lawyers and clients identify their non-legal interests (31%); mediator has reputation for settling cases (28%); mediator has experience as a judge (25%); mediator is good at "knocking heads" (16%)). But see Susan Raines, et al., Best Practicesfor Mediation Training and Regulation: PreliminaryFindings, 48 136 Harvard Negotiation Law Review [Vol. 18:71 certified or admitted to a roster. 238 A substantial number of courts require their mediators to be lawyers, often with a specified number of years of experience. Interestingly, research fails to confirm that subject matter expertise results in a greater number of settlements.239 On the other hand, research indicates that settlement is 240 more likely if mediators are trusted and viewed as "high quality," and, as noted supra, when lawyers are permitted to play a role in selecting mediators, they generally seem to prefer other lawyers with relevant subject matter expertise. Some of the most experienced court programs use a mix of these various requirements, tailored to particular practice areas. 241 Meanwhile, some ADR organizations have found ways to facilitate potential clients' ability to assess mediators' reputations and approaches. The International Mediation Institute, for example, posts information gathered from former cli- 242 ents regarding their experience with mediators. Many courts have also established ethical rules for the mediators who handle court referrals. These rules represent a means to hold the mediators accountable to the court. Less obvious but just as important, such rules and their transparent enforcement help to make

FAm. CT. REV. 541, 544-45 (2010) (observing that authors used a collaborative assessment process that involved a broad range of stakeholders, including Florida judges, court program administrators, mediation trainers, mediators and DRC staff, in order to make recommendations for "best practices" in mediation training). 238. See, e.g., FLA. STAT. § 10.100 (West 2012) (describing a multitude of requirements that must be fulfilled in order to gain court approval to serve as a mediator). 239. See Wissler, Court-ConnectedMediation, supra note 90, at 699. 240. See, e.g., Douglas Henderson, Mediation Success: An Empirical Analysis, 11 OHIO ST. J. ON Dirp. RESOL. 105, 132-45 (1996) (focus on construction mediation; "high quality" not defined); see also Stephen B. Goldberg & Margaret L. Shaw, FurtherIn- vestigation into the Secrets of Successful and Unsuccessful Mediators, 26 ALT. TO HIGH COST LITIG. 149 (2008) ("The sole characteristic shared by nearly all the 13 mediators in the top half on the overall scores was that 11 of the 13 were a standard deviation above the mean on at least one of the confidence-building attributes"). 241. See Welsh, supra note 226, at 496-97 (describing Florida's system for certify- ing mediators); McAdoo & Welsh, supra note 117, at 19-21. 242. See Irena Vanenkova, What's in it for me? Certification and the Laws of Sup- ply and Demand, 5 N.Y. Disp. Resol. Lawyer 51, 52 (describing certification as conducted by the International Mediation Institute, including third party's solicitation and distillation of feedback from clients for posting online); see, e.g., Feedback Exam- ple, INTERNATIONAL MEDIATION INSTITUTE, http://imimediation.org/feedback-example (last visited Oct. 28, 2012); see also Michael McIlwrath & Rolan Schroeder, Trans- parency in International Arbitration: What Are Arbitrators and Institutions Afraid Of?, in CONTEMPoRARY IssuEs 2010, supra note 17, at 343-44, 347-49, 351 (proposing information that should be made available by arbitral institutions and arbitrators to assist with parties' selection of arbitrators, decrying the reliability of word-of-mouth recommendations or ad hoc, individual feedback to arbitrators, and referencing a standardized arbitrator evaluation form developed in 2006 at the behest of Thomas Wolde). Spring 2013] Bilateral Investment Treaty Arbitration the courts accountable to the parties who participate in court-connected mediation processes and to the public more generally. Courts' ethical rules are based largely on the Model Standards of Conduct for Mediators adopted by the ABA, AAA, and SPIDR (later ACR) in 1994 and 2005.243 But relatively few courts have established formal grievance or performance evaluation procedures. 244 Even fewer courts have established mechanisms for the provision of advisory opinions to guide their mediators. 245 These formal and advisory mechanisms, where they exist and are transparent, help to ensure that mediators are truly accountable to the courts and that courts themselves are 246 accountable to the parties and public.

3. Accountability of the Courts to the Parties and the Public The final method that courts use to assure the quality of their mediation programs (and to demonstrate both their own and their mediators' accountability) involves post-mediation evaluation. 247 Af- ter mediation sessions have occurred, some courts solicit and collect

243. MODEL STANDARDS OF CONDUCT FOR MEDIATORS (1994); see also MODEL STAN. DARDS OF CONDUCT FOR MEDIATORS (2005); R. Wayne Thorpe & Susan M. Yates, An Overview of the Revised Model Standards of Conduct for Mediators, 12 Disp. RESOL. MAG. 30, (2005-2006); Michael L. Moffitt, The Wrong Model, Again: Why the Devil Is Not in the Details, 12 Disp. RESOL. MAG. 31, (2005-2006); Joseph B. Stulberg, The Model Standards of Conduct: A Reply to Professor Moffitt, 12 Disp. RESOL. MAG. 34, (2005-2006). 244. See McAdoo & Welsh, supra note 117, at 24-25 (regarding support and evaluation); Riskin & Welsh supra note 113, at 927. The ABA Dispute Resolution Section recently adopted a policy supporting credentialing of mediators provided that credentialing programs: clearly define the skills, knowledge and values which persons it cre- dentials must possess; ensure that candidates have training adequate to instill those skills, knowledge and values; be administered by an organization distinct from the organization which trains the candidate; have an assessment process capable of determining with consistency whether or not candidates possess the defined skills, knowledge and values; explain clearly to persons likely to rely on the credential what is being certified; and provide an accessible, transparent system to register complaints against credentialed mediators, promptly and fairly investigate complaints and, if appropriate, de-credential a mediator who fails to comply with standards. A majority of the Task Force that recommended this policy also believed that organizations should have a process to monitor the performance of credentialed mediators, such as periodic requests for feedback. See ALTERNATIVE DISPUTE RESOLUTION SECTION OF THE AMERI- CAN BAR ASSOCIATION, TASK FORCE ON MEDIATOR CREDENTIALING: FINAL REPORT (Au- gust 2012). 245. See McAdoo & Welsh, supra note 117, at 12-13. 246. See Sharon Press, Institutionalizationof Mediation in Florida:At the Cross- roads, 108 PENN ST. L. REV. 43, 53-55 (2003); Sharon Press, Florida Provides a Sys- tem of Grievances Against Mediators, 7 DIsP. RESOL. MAG. 8, (2001). 247. See Roselle L. Wissler & Robert W. Rack, Jr., Assessing Mediator Perform- ance: The Usefulness of ParticipantQuestionnaires, 2004 J. Disp. RESOL. 229 (2004) (assessing the usefulness of participant assessments of mediator performance); Harvard Negotiation Law Review [Vol. 18:71

evaluations from the parties and their attorneys. Very frequently, these surveys include questions designed to elicit perceptions regarding mediation's effectiveness, efficiency, procedural justice, and distributive justice. Parties are much less likely than their lawyers to complete these surveys unless they are administered and collected immediately after the mediation session. This difference in the likelihood of receiving parties' feedback is significant because research results suggest divergence between lawyers' perceptions and those of their clients. 248 Courts' use of these surveys also varies. Courts frequently do not have the resources required to conduct a thorough review of the data they have collected and to consider what the data may mean. A smaller number of courts keep quantitative data regarding the number of cases referred to mediation, the types of cases, key events, the date of disposition, and mediation outcomes. Unfor- tunately, federal and state courts, even courts located in the same state, record different pieces of data, making comparisons difficult.249 In some infrequent cases, courts do invest the resources required for rigorous quantitative evaluation 250 or for staff members to conduct

Wayne D. Brazil, Hosting Mediations as a Representative of the System of Civil Jus- tice, 22 OHIO ST. J. ON Disp. RESOL. 227, 251-253 (2007) (describing the post-mediation questionnaires distributed by the U.S. District Court of the Northern District of California). 248. See TAMARA RELIS, PERCEPTIONS IN LITIGATION AND MEDIATION: LAWYERS, DE- FENDANTS, PLAINTIFFS, AND GENDERED PARTIES 33-4 (2009); Tamara Relis, It's Not About the Money!: A Theory on Misconceptions of Plaintiffs'Litigation, 68 U. PITT. L. REV. 701, 728-33 (2007); Schneider & Fleury, supra note 201, at 389-90; Riskin & Welsh, supra note 113, at 910 (noting divergence of perceptions and preferences between lawyers and many, especially one-shot player, clients); Welsh, Making Deals, supra note 83, at 842-44; Welsh, Thinning Vision, supra note 112, at 4-5. 249. See Memorandum from ABA Section of Dispute Resolution Task Force on Re- search and Statistics to Court Administratorsand ADR ProgramAdministrators, Top Ten Pieces of Information Courts Should Collect on ADR 1 (2005) (recognizing constraints on courts' information management systems but requesting collection of ADR-related information); see also Court ADR Instruction Manual, RSI, http:// courtadr.org/manual/monitor.php (last visited Oct. 28, 2012). 250. See, e.g., JAMES S. KAKALIK, ET AL., JUST, SPEEDY AND INEXPENSIVE? AN EVAL- UATION OF JUDICIAL CASE MANAGEMENT UNDER THE CIL JUSTICE REFORM ACT (1996); JAMES S. KAKALIK, ET AL., AN EVALUATION OF MEDIATION AND EARLY NEUTRAL EVALU- ATION UNDER THE CIVIL JUSTICE REFORM ACT (1996); STIENSTRA ET AL., supra note 215, at i-iii; Wayne Kobbervig, Mediation of Civil Cases in Hennepin County: An Evaluation 23-26 (Minn. Judicial Center 1991); Anderson & Pi, supra note 184, at 566 (California data). There is some controversy about whether a quantitative study can be considered sufficiently valid if it does not involve random assignment and control groups. Needless to say, courts are likely to find it very difficult to prioritize research purity as more important than providing people with access to a process they believe to be beneficial. Researchers, as well, must struggle with this issue of access. See The Spring 2013] Bilateral Investment Treaty Arbitration observations of mediators. 2 51 Certain agencies, particularly the U.S. Postal Service, have affirmatively selected particular models of mediation and have then implemented monitoring and evaluation systems 252 to ensure their use. Obviously, engaging in hands-on efforts to ensure the quality of mediators and mediations requires funding and staffing. Many courts have adopted mediation programs for the exact purpose of saving time and costs. Investing in the staffing and evaluation that is required for meaningful quality control thus may seem counter-intui- tive. In some states, though, such evaluation is done on a regular basis for the other key neutrals located in courts, e.g., judges, magis- 253 trate judges, etc., in order to provide these neutrals with feedback. Though such evaluations are not generally shared with the public, some courts have recently shared data as they have assessed the pub- 25 4 lic's general perceptions of the courts.

4. Recommendation for the Investor-State Context Although it is worthwhile to consider the experience of U.S. courts and agencies in assuring quality and accountability, it is also important to recognize that the investment treaty context involves many fewer cases and may be even more likely to be dominated by a small group of repeat-player lawyers who represent both investors and nations. Given this difference, the best means of assuring the quality of mediation, at least initially, is likely to lie in selecting mediators who are already trusted by the repeat players.2 55 For

National Commission for the Protection of Human Subjects of Biomedical and Behav- ioral Research, The Belmont Report: Ethical Principles and Guidelines for the Protec- tion of Human Subjects of Research, 45 C.F.R. § 46 (1979) (listing the "basic ethical principles" as respect for persons, beneficence and justice and observing that "[an injustice occurs when some benefit to which a person is entitled is denied without good reason or when some burden is imposed unduly"), available at http:// www.hhs.gov/ohrp/humansubjects/guidance/belmont.html. 251. See Brad Honoroffet al., Putting Mediation Skills to the Test, 6 NEGOT. J. 37, 228-38 (1990) (describing experiment in Massachusetts; note, however, that there is no longer widespread court-connected non-family civil mediation in Massachusetts); see also Grace D'Alo, Accountability in Special Education Mediation: Many a Slip 'Twixt Vision and Practice?, 8 HARV. NEGOT. L. REV. 201, 229-37 (2003) (describing observations of special education mediators). 252. See Nabatchi & Blomgren Bingham, supra note 142, at 405-409 (2001). 253. See CourTools, supra note 110, at 2 (regarding procedural justice-related feedback provided to courts and individual judges). 254. See ROrrMAN, supra note 86, at 1. 255. This cannot be the only safeguard, of course. See e.g., Rogers, supra note 54, Gottwald, supra note 54; James R. Coben, Creatinga 21st Century Oligarchy: Judi- cial Abdication to Class Action Mediators, 5 PENN ST. Y.B. ON ARB & MEDIATION - (forthcoming 2013) (expressing concerns regarding reliance on small pool of mediators Harvard Negotiation Law Review [Vol. 18:71

many years, for example, a small group of trusted men served as international commercial arbitrators. 256 The prestige earned by the arbitration process was due, at least in part, to these men's involvement in it.257 Current investment treaty disputants could rely upon the breadth of experience and good judgment-social, legal, and political-of this group of men and women,258 on a temporary basis. Mechanisms to bring these mediators together on a regular basis would have the potential to encourage their creation of a shared vision, body of substantive knowledge, skills, and sense of identity and ethics.259 Collectively, they may also develop the model (or models 260 ) of mediation that could serve as the future defaults in this context. If this approach to mediator selection is taken, however, "succes- sion planning" would also have to be done, quickly, with resources identified to permit observation of these mediators and mentoring of potential mediators whose professional backgrounds, nationalities, ethnicities and gender will signal to all stakeholders-not just the

to provide evidence that class action settlements are the result of arms-length negotiation and not the result of collusion). 256. See Yves Dezalay & Bryant Garth, Merchants of Law as Moral Entrepre- neurs: Constructing International Justice from the Competition for Transnational Business Disputes, 29 LAw & Soc'y REV. 27, 30-33 (1995). Interestingly, it appears that parties continue to use a relatively small number of individuals for international arbitration. See, e.g., McIlwrath & Schroeder, supra note 242, at 355-56 (noting that the current system of international arbitration leads to repeated use of same arbitrators and perception of "arbitration 'mafia'"); Elsing, supra note 59, at 308 (raising concerns regarding lack of access to international arbitrators). 257. Such men likely gained their persuasive power from both their pre-existing status and their performance in individual cases. See Donna Shestowsky, Psychology and Persuasion, in THE NEGOTIATOR'S FIELDBOOK, supra note 83, at 361-70. 258. Note, however, that some commentators have expressed uncertainty regarding the ability to identify those who should serve as investment treaty mediators, observing that the appropriate expertise "requires understanding of public policy dimensions which are grafted onto purely commercial interests." Cheng, supra note 17, at 440. Others have worried particularly that moving international commercial arbitrators into investor-state arbitration has resulted in a structural bias favoring investors over states. David Schneiderman, Judicial Politics and International In- vestment Arbitration: Seeking An Explanation for Conflicting Outcomes, 30 NW. J. INT'L L. & Bus. 383 (2010). This concern would need to be addressed in mediator selection as well. 259. See Welsh, Institutionalizationand Professionalization,supra note 226, at 494-5 (describing the elements of professionalism); Craig McEwen, Giving Meaning to Mediator Professionalism, 11 DisP.RESOL. MAG. 3, 5 (2005) (regarding value of "active communities of practice"). 260. Recalling, of course, there will and must be process pluralism-which some colleagues seem to recall almost effortlessly. See, e.g., Carrie Menkel-Meadow, supra note 77, at 555; John Lande, supra note 77, at 147. Spring 20131 Bilateral Investment Treaty Arbitration repeat players-that they will receive a procedurally just process. 26 1 In other words, the reputations and identities of the mediators will need to assure stakeholders that they can trust the process and will be allowed to express themselves, will receive sincere and trustworthy consideration, and will be treated in an even-handed and respectful manner in a neutral forum. 262 These signals will matter. At the same time, of course, results should be transparent in order to permit stakeholders to assess whether there is an emerging and consistent pattern of bias.263 As described supra, and despite the stature of these mediators, the investment mediation process should include both a grievance procedure and periodic evaluation of these mediators.264 Indeed, mediators of this stature are likely to be ready to model compliance with such procedures. Finally, the institutions managing mediation in the investor- state context should mimic the transparency modeled by those court,

261. On this point, Professor Mariana Hernandez Crespo has recently proposed the use of co-mediation. 262. Welsh, Embedded Neutrals, supra note 112, at 424-25; see Jill I. Gross, SEC Seeks Public Comment on Two FINRA Dispute Resolution Rule Proposals,ADR PROF. BLOG (Mar. 4, 2012), http://www.indisputably.org/?p=3445; Jill I. Gross, SEC Ap- proves FINRA Proposal to Exempt Collective Actions from FINRA Arbitration, ADR PROF BLOG (Apr. 22, 2012) http://www.indisputably.org/?p=3558; Jill I. Gross, SEC Approves Rule Change to Raise the Cap for FINRA Simplified Arbitration Claims to $50,000, ADR PROF BLOG (May 5, 2012) http://www.indisputably.org/?p=3592. 263. We realize, of course, that such a call for transparency may conflict with parties' desire for confidential mediated settlement agreements. Some magistrate judges in the federal district courts in the U.S. have also dealt with this issue by creating "settlement logs" that permit the magistrates to speak with some knowledge regarding the range of settlements for particular types of disputes, without revealing confidential identifying information. See Morton Denlow & Jennifer E. Shack, Judicial Settlement Databases:Development and Uses, 43 JUDGES J. 19, 19 (2004). Perhaps this innovation could be used (or adapted) by ICSID, UNCTAD and other providers of dispute resolution in the investment treaty context. For the development of greater transparency in the NAFTA context, see Coe, Transparency,supra note 7, at 1339-40. Interestingly, even though court-connected research reveals that the public's approval of courts is likely to be improved by reporting regarding courts' performance, such reports appear unlikely to affect their perceptions of the procedural fairness provided by the courts. See RorrMAN, supra note 86, at 35. 264. Writing about how the two ways in which organizations gain deference, Tom Tyler has observed: An organization can gain deference by having formal rules that reflect neutrality. It can also gain deference through the personal relationships that exist between employees and their own particular supervisors. The former approach reflects a neutrality model of procedural fairness, the latter approach a trust-based model. Similarly, the police can gain deference because they are viewed as following professional rules of conduct and uniform procedures, or particular police officers can be respected and known in their communities and can, through these personalized connections, gain deference. Tyler, Social Justice, supra note 83, at 122. Harvard Negotiation Law Review [Vol. 18:71

agency and non-profit organizations that conduct regular process evaluations, collate and disseminate the aggregate results, and make responsive changes. 265 These management choices, too, should enhance trust-and provoke even more additional, productive 6 mimicry. 26

V. CONCLUSION The viability of investor-state arbitration is dependent upon states' and investors' faith that the process will help them achieve their complementary goals of increased private investment, the avoidance of unfair expropriation, and the provision of a forum for the prompt and fair resolution of disputes. Some states' refusal to comply with arbitration awards, as well as some investors' pursuit of creative claims, threatens to undermine this necessary element of faith and, with it, the regime of investor-state investment treaties. Meanwhile, the expense and contentiousness of the arbitration process itself has become an impediment to its use. If arbitration is to play its valuable role in facilitating international investment, it requires the addition of other processes and the development of a coherent dispute resolution system. Dispute system design principles clearly counsel the value of various process choices; the most robust systems offer more than one process to meet the needs of stakeholders and intentionally incorporate processes that focus presumptively on interests as distinct from rights. Procedural justice theories and research should also guide the development or amendment of dispute resolution systems, with reminders to provide stakeholders the opportunity to speak, reassurance that their views

265. See Welsh, Embedded Neutrals, supra note 112 at 421, 423-24 (describing researcher's use of data that is regularly collected in California regarding arbitration, as well as FINRA's use of evaluation data to make procedural changes); Sandra K. Partridge, Arbitration Post-AT&T Mobility v. Concepcion at the American Arbitration Association-A Service Provider's Perspective, 4 PENN ST. Y.B. ON ARB. & MEDIATION 81, 83-84, 86 (2013) (describing searchable online docket to track cases through each stage of the AAA Class Arbitration process, as well as data that AAA makes available regarding the consumer cases that it administers); Alexander J.S. Colvin, Empirical Research on Employment Arbitration:Clarity Amidst the Sound and Fury, 11 EMP. RTS. & EMP. POL'Y J. 405 (2007) (describing the data on arbitration that is publicly available in California); see also CPR-GEORGETOWN COMMISSION ON ETHICS AND STAN- DARDS OF PRACTICE IN ADR, PRINCIPLES FOR ADR PROVIDER ORGANIZATIONS 17 (2002); Nancy H. Rogers, et al., DESIGNING SYSTEMS AND PROCESSES FOR MANAGING DisPuTEs 319-356 (2013) (providing an overview of the systematic collection of data, and the use of such data, for those designing and revising dispute systems). 266. See Riskin & Welsh, supra note 113, at 912 (urging private ADR providers to adopt court-focused proposals "for competitive or altruistic reasons, or both"); see also Barendrecht, supra note 175, at 14-20. Spring 2013] Bilateral Investment Treaty Arbitration will be considered, a neutral forum, and treatment that is even- handed and dignified. The importance of these procedural markers can appear obvious, but decision-makers frequently lose sight of them when faced with impending deadlines, potentially dire circumstances, and no guarantee that procedurally just procedures will produce cheerful consensus. Adding mediation to the options already used on a regular basis in investment treaty disputes has the potential to permit earlier resolution of disputes, improve communication and information ex- changes, enhance mutual understanding, encourage explicit consideration of parties' interests as well as relevant rights, and produce integrative solutions and parties who understand that they par- ticipated in a "justice event." This Article has made clear, however, that the value of mediation and the particular model that is most appropriate will depend upon the other processes available in the dispute resolution system. The different names given to the processes- i.e., "mediation" vs. "conciliation" vs. "arbitration"-will not be enough to distinguish among them. It will be essential to dig more deeply into the actual functions of the other processes that exist in the dispute resolution system. Further, both dispute system design principles and procedural justice theory and research urge that representatives of investors and other affected stakeholders should be involved in the decision-making process that produces the dispute resolution system incorporated into the dispute resolution clauses in treaties. Indeed, the input of such stakeholders should be sought regarding the addition of mediation to such a system, the model or models that will be used, appropriate referral mechanisms, and quality controls. Based on procedural justice research, this Article suggests that the default mediation process should begin in a facilitative manner in order to increase the likelihood of trust-building and information exchange. Such trust-building and information exchange is most likely to yield information regarding disputing states' and investors' underlying interests and create the potential for the development of integrative solutions. At the same time, this model should only be presumptive and should also permit both evaluative interventions by the mediator and discussion of relevant legal norms. Thus, the process itself will provide opportunity for the "loop forwards" and "loop backs" recommended by dispute system design principles. Particularly if stakeholders' input is sought and considered regarding mechanisms for the referral of disputes to mediation, procedural justice research further suggests that mediation, or a more 144 Harvard Negotiation Law Review [Vol. 18:71 limited element of the process, could be made compulsory. As this Article has detailed, courts in the U.S. have experimented with a wide variety of options. The requirement of compulsory participation in an initial meeting would permit the parties themselves to choose whether to proceed further in mediation at that time or specify the timing of its future use. The parties' engagement in such decision- making would be consistent with both dispute system design principles and likely produce advantageous perceptions of procedural justice. Finally, this Article has introduced the variety of quality assurance and accountability measures that may be undertaken if mediation is integrated into the investment treaty context. Such quality assurance and accountability measures will be essential. In the short term, the establishment of a small pool of well-known and well- respected investment treaty mediators might offer a reasonably strong and pragmatic guarantee of quality and engender a heightened perception of trust in the process. In the long term, evaluation and mentoring should be put into place to permit thoughtful cultivation of both the model of mediation that is best suited for the investor-state context and the next generation of mediators. The ultimate goal of investment treaty dispute resolution is to foster foreign investment in a manner that is fair for investors seeking profitable opportunities and for states working toward the common good of their people. Balanced appropriately, investors' and states' interests should result in a mutually-reinforcing relationship. Implemented in a manner consistent with dispute system design and procedural justice, mediation may help investors and states remem- ber why they were attracted to each other in the first place and whether today's troubles simply signal the need to move to the next, more realistic stage of their relationship. This is not as exciting, perhaps, as the fireworks of a hard-fought arbitration, but fireworks are costly and best saved for special occasions. With some well-timed and well-executed help, the flame of mediation may cast a more modest but more sustained light, to allow investors and states to join together in finding a path through the darkness.